U.S. patent application number 16/418757 was filed with the patent office on 2019-11-21 for computational model management.
This patent application is currently assigned to State Street Corporation. The applicant listed for this patent is State Street Corporation. Invention is credited to Dushyant Ralhan.
Application Number | 20190354809 16/418757 |
Document ID | / |
Family ID | 68533609 |
Filed Date | 2019-11-21 |
![](/patent/app/20190354809/US20190354809A1-20191121-D00000.png)
![](/patent/app/20190354809/US20190354809A1-20191121-D00001.png)
![](/patent/app/20190354809/US20190354809A1-20191121-D00002.png)
![](/patent/app/20190354809/US20190354809A1-20191121-D00003.png)
![](/patent/app/20190354809/US20190354809A1-20191121-D00004.png)
![](/patent/app/20190354809/US20190354809A1-20191121-D00005.png)
![](/patent/app/20190354809/US20190354809A1-20191121-D00006.png)
![](/patent/app/20190354809/US20190354809A1-20191121-D00007.png)
![](/patent/app/20190354809/US20190354809A1-20191121-D00008.png)
![](/patent/app/20190354809/US20190354809A1-20191121-D00009.png)
![](/patent/app/20190354809/US20190354809A1-20191121-D00010.png)
View All Diagrams
United States Patent
Application |
20190354809 |
Kind Code |
A1 |
Ralhan; Dushyant |
November 21, 2019 |
COMPUTATIONAL MODEL MANAGEMENT
Abstract
Systems, methods, apparatuses, and computer-readable media for
computational model management are described. In one embodiment, an
apparatus may include a processor and a memory storing
instructions, which, when executed by the processor, cause the
processor to generate a candidate computational model for a model
function based on an objective and function data, and provide the
candidate computational model to a model assessment process to
perform a champion-challenger process based on performance metrics,
industrial metrics, and user metrics. Other embodiments are
described.
Inventors: |
Ralhan; Dushyant; (Boston,
MA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
State Street Corporation |
Boston |
MA |
US |
|
|
Assignee: |
State Street Corporation
Boston
MA
|
Family ID: |
68533609 |
Appl. No.: |
16/418757 |
Filed: |
May 21, 2019 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
62674381 |
May 21, 2018 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G06K 9/6264 20130101;
G06F 40/151 20200101; G06K 9/6257 20130101; G06F 40/289 20200101;
G06F 16/116 20190101; G06F 40/186 20200101; G06F 40/30 20200101;
G06F 16/90332 20190101; G06F 40/131 20200101; G06F 40/216 20200101;
G06N 20/00 20190101; G06F 40/279 20200101; G06N 3/0454 20130101;
G06F 16/93 20190101; G06F 16/906 20190101; G06F 16/13 20190101;
G06F 16/353 20190101; G06K 9/6218 20130101; G06K 9/623 20130101;
G06F 40/295 20200101 |
International
Class: |
G06K 9/62 20060101
G06K009/62; G06N 3/04 20060101 G06N003/04 |
Claims
1. An apparatus, comprising: a processor; and a memory storing
instructions which when executed by the processor cause the
processor to: generate a candidate computational model for a model
function based on an objective and function data, and provide the
candidate computational model to a model assessment process to
perform a champion-challenger process based on performance metrics,
industrial metrics, and user metrics.
2. The apparatus of claim 1, the instructions, when executed by the
processor, to cause the processor to perform a model decay process
to determine whether a computational model has decayed.
3. The apparatus of claim 1, the instructions, when executed by the
processor, to cause the processor to create a working set for
generating the candidate computational model.
4. The apparatus of claim 3, the working set comprising a data
version topic and a model version topic.
5. The apparatus of claim 3, the instructions, when executed by the
processor, to cause the processor to provide access control to the
working set on a per-topic level.
6. The apparatus of claim 3, the instructions, when executed by the
processor, to cause the processor to provide multiple different
versions of the working set.
7. The apparatus of claim 3, the instructions, when executed by the
processor, to cause the processor to generate the working set for a
business case.
8. A method, comprising: generating a candidate computational model
for a model function based on an objective and function data; and
providing the candidate computational model to a model assessment
process to perform a champion-challenger process based on
performance metrics, industrial metrics, and user metrics.
9. The method of claim 8, comprising performing a model decay
process to determine whether a computational model has decayed.
10. The method of claim 8, comprising creating a working set for
generating the candidate computational model.
11. The method of claim 10, the working set comprising a data
version topic and a model version topic.
12. The method of claim 10, comprising providing access control to
the working set on a per-topic level.
13. The method of claim 10, comprising providing multiple different
versions of the working set.
14. The method of claim 10, comprising generating the working set
for a business case.
15. A non-transitory computer-readable storage medium storing
instructions that when executed by a processor circuit cause the
processor circuit to: generate a candidate computational model for
a model function based on an objective and function data; and
provide the candidate computational model to a model assessment
process to perform a champion-challenger process based on
performance metrics, industrial metrics, and user metrics.
16. The computer-readable storage medium of claim 15, storing
instructions that when executed by the processor circuit cause the
processor circuit to performing a model decay process to determine
whether a computational model has decayed.
17. The computer-readable storage medium of claim 15, storing
instructions that when executed by the processor circuit cause the
processor circuit to creating a working set for generating the
candidate computational model.
18. The computer-readable storage medium of claim 17, the working
set comprising a data version topic and a model version topic.
19. The computer-readable storage medium of claim 17, storing
instructions that when executed by the processor circuit cause the
processor circuit to providing access control to the working set on
a per-topic level.
20. The computer-readable storage medium of claim 17, storing
instructions that when executed by the processor circuit cause the
processor circuit to providing multiple different versions of the
working set.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims the benefit of U.S. Provisional
Application No. 62/674,381 filed May 21, 2018. The aforementioned
provisional application is incorporated by reference herein in its
entirety.
TECHNICAL FIELD
[0002] Embodiments disclosed herein relate to the field of
computational models. More specifically, embodiments disclosed
herein relate to cataloging and evaluating multiple versions of
computational models.
BACKGROUND
[0003] Computational model management requires interoperation
between several individual functions including, for example, data
acquisition, persistence, modelling experiments, deployment,
versioning, release, and/or the like. In addition, a lack of
reference standards, unproven patterns, and/or convoluted workflows
combined with an overload of machine learning (ML) tools and
frameworks further complicates the process. Accordingly,
coordinating the results of these disparate platforms to manage
existing and/or determine new computational models using
conventional systems is extremely complex and inefficient.
BRIEF DESCRIPTION OF THE DRAWINGS
[0004] FIG. 1 depicts hardware components of a data transmission
network.
[0005] FIG. 2 depicts an example network including an example set
of devices communicating with each other over an exchange
system.
[0006] FIG. 3 illustrates an embodiment of an example system.
[0007] FIG. 4 illustrates an embodiment of exemplary logic and data
flows through a computing architecture.
[0008] FIG. 5 illustrates an embodiment of a first logic flow.
[0009] FIG. 6 illustrates an embodiment of a neural network.
[0010] FIG. 7 illustrates an embodiment of an exemplary computing
architecture.
[0011] FIG. 8 illustrates a first operating environment according
to some embodiments.
[0012] FIG. 9 illustrates a second operating environment according
to some embodiments.
[0013] FIG. 10 illustrates a third operating environment according
to some embodiments.
[0014] FIG. 11 illustrates a fourth operating environment according
to some embodiments.
[0015] FIG. 12 illustrates a fifth operating environment according
to some embodiments.
[0016] FIG. 13 illustrates a sixth operating environment according
to some embodiments.
[0017] FIG. 14 illustrates an embodiment of a second logic
flow.
[0018] FIG. 15 illustrates an embodiment of a third logic flow.
[0019] FIG. 16 illustrates an embodiment of a fourth logic
flow.
[0020] FIG. 17 illustrates an embodiment of a fifth logic flow.
[0021] FIG. 18 illustrates an embodiment of a sixth logic flow.
[0022] FIG. 19 illustrates an embodiment of a seventh logic
flow.
[0023] FIG. 20 illustrates an embodiment of an eighth logic
flow.
[0024] FIG. 21 illustrates an embodiment of a ninth logic flow.
[0025] FIG. 22 illustrates an embodiment of a tenth logic flow.
[0026] FIG. 23 illustrates an embodiment of an eleventh logic
flow.
DETAILED DESCRIPTION
[0027] Embodiments disclosed herein provide techniques, processes,
and/or apparatus to manage computational models, including, without
limitation, cataloguing and persisting existing models, and/or
evaluating, releasing, and/or deploying new models. For example, in
some embodiments, a computational model process may operate to
catalog and persist multiple versions of models and data,
including, for instance, their access control and meta information.
In another example, some embodiments may provide a computational
model process operative to score and/or compare model performance
using various categories of metrics for model evaluation (for
example, industrial metrics, user-defined metrics, and/or the
like). In various embodiments, a computational model process may
operate to perform certain corrective actions to address model
deterioration, such as via model decay, by operationally deploying
improved model versions, for example, using approvals and
deployment automation.
[0028] In various embodiments, a computational model process may
operate to manage a machine learning (ML) computational model
lifecycle. In various embodiments, the models may include various
computational models, including, without limitation, artificial
intelligence (AI) frameworks, neural network models, and/or the
like. Non-limiting examples of a computational model may include an
ML model, an artificial intelligence (AI) model, a neural network
(NN), an artificial neural network (ANN), a convolutional neural
networks (CNN), a deep learning (DL) network, a deep neural network
(DNN), a recurrent neural network (RNNs), combinations thereof,
variations thereof, and/or the like. Embodiments are not limited in
this context.
[0029] ML development using conventional techniques is complex and
inefficient, making it difficult for developers to achieve useful
ML models within a sufficient time frame. For example, ML
algorithms have myriads of configurable parameters such that it is
difficult to track which parameters, code, and data went into each
experiment to produce a model. Even with detailed tracking
platforms, it is challenging to achieve reproducible results and to
get the same ML code to work the same way repeatedly. Deploying a
model to production may be challenging due to the different types
and configurations of deployment tools and operating environments.
Accordingly, ML development using conventional techniques is not as
robust, predictable, and, therefore, widespread as traditional
software development processes.
[0030] Accordingly, some embodiments may include computational
model processes operative to provide a computational model
lifecycle that is more efficient, robust, and effective than
conventional techniques.
[0031] In this description, numerous specific details, such as
component and system configurations, may be set forth in order to
provide a more thorough understanding of the described embodiments.
It will be appreciated, however, by one skilled in the art, that
the described embodiments may be practiced without such specific
details. Additionally, some well-known structures, elements, and
other features have not been shown in detail, to avoid
unnecessarily obscuring the described embodiments.
[0032] In the following description, references to "one
embodiment," "an embodiment," "example embodiment," "various
embodiments," etc., indicate that the embodiment(s) of the
technology so described may include particular features,
structures, or characteristics, but more than one embodiment may
and not every embodiment necessarily does include the particular
features, structures, or characteristics. Further, some embodiments
may have some, all, or none of the features described for other
embodiments.
[0033] As used in this description and the claims and unless
otherwise specified, the use of the ordinal adjectives "first,"
"second," "third," etc. to describe an element merely indicate that
a particular instance of an element or different instances of like
elements are being referred to, and is not intended to imply that
the elements so described must be in a particular sequence, either
temporally, spatially, in ranking, or in any other manner.
[0034] FIG. 1 is a block diagram that provides an illustration of
the hardware components of a data transmission network 100,
according to embodiments of the present technology. Data
transmission network 100 is a specialized computer system that may
be used for processing large amounts of data where a large number
of computer processing cycles are required.
[0035] Data transmission network 100 may also include computing
environment 114. Computing environment 114 may be a specialized
computer or other machine that processes the data received within
the data transmission network 100. Data transmission network 100
also includes one or more network devices 102. Network devices 102
may include client devices that are capable of communicating with
computing environment 114. For example, network devices 102 may
send data to the computing environment 114 to be processed, may
send signals to the computing environment 114 to control different
aspects of the computing environment or the data it is processing,
among other reasons. Network devices 102 may interact with the
computing environment 114 through a number of ways, such as, for
example, over one or more networks 108. As shown in FIG. 1,
computing environment 114 may include one or more other systems.
For example, computing environment 114 may include a database
system 118 and/or a communications grid 120.
[0036] In other embodiments, network devices may provide a large
amount of data, either all at once or streaming over a period of
time to the computing environment 114 via networks 108. For
example, network devices 102 may include network computers,
sensors, databases, or other devices that may transmit or otherwise
provide data to computing environment 114. For example, network
devices may include local area network devices, such as routers,
hubs, switches, or other computer networking devices. These devices
may provide a variety of stored or generated data, such as network
data or data specific to the network devices themselves. Network
devices may also include sensors that monitor their environment or
other devices to collect data regarding that environment or those
devices, and such network devices may provide data they collect
over time. Network devices may also include devices within the
internet of things (IoT), such as devices within a home automation
network. Some of these devices may be referred to as edge devices
and may involve edge computing circuitry. Data may be transmitted
by network devices directly to computing environment 114 or to
network-attached data stores, such as network-attached data stores
110 for storage so that the data may be retrieved later by the
computing environment 114 or other portions of data transmission
network 100.
[0037] Data transmission network 100 may also include one or more
network-attached data stores 110. Network-attached data stores 110
are used to store data to be processed by the computing environment
114 as well as any intermediate or final data generated by the
computing system in non-volatile memory. However, in certain
embodiments, the configuration of the computing environment 114
allows its operations to be performed such that intermediate and
final data results can be stored solely in volatile memory (e.g.,
RAM), without a requirement that intermediate or final data results
be stored to non-volatile types of memory (e.g., disk). This can be
useful in certain situations, such as when the computing
environment 114 receives ad hoc queries from a user and when
responses, which are generated by processing large amounts of data,
need to be generated on-the-fly. In this non-limiting situation,
the computing environment 114 may be configured to retain the
processed information within memory so that responses can be
generated for the user at different levels of detail as well as
allow a user to interactively query against this information.
[0038] Network-attached data stores 110 may store a variety of
different types of data organized in a variety of different ways
and from a variety of different sources. For example,
network-attached data storage may include storage other than
primary storage located within computing environment 114 that is
directly accessible by processors located therein. Network-attached
data storage may include secondary, tertiary or auxiliary storage,
such as large hard drives, servers, virtual memory, among other
types. Storage devices may include portable or non-portable storage
devices, optical storage devices, and various other mediums capable
of storing, containing data. A machine-readable storage medium or
computer-readable storage medium may include a non-transitory
medium in which data can be stored and that does not include
carrier waves and/or transitory electronic signals. Examples of a
non-transitory medium may include, for example, a magnetic disk or
tape, optical storage media such as compact disk or digital
versatile disk, flash memory, memory or memory devices. A
computer-program product may include code and/or machine-executable
instructions that may represent a procedure, a function, a
subprogram, a program, a routine, a subroutine, a module, a
software package, a class, or any combination of instructions, data
structures, or program statements. A code segment may be coupled to
another code segment or a hardware circuit by passing and/or
receiving information, data, arguments, parameters, or memory
contents. Information, arguments, parameters, data, etc. may be
passed, forwarded, or transmitted via any suitable means including
memory sharing, message passing, token passing, network
transmission, among others. Furthermore, the data stores may hold a
variety of different types of data. For example, network-attached
data stores 110 may hold unstructured (e.g., raw) data, such as
manufacturing data (e.g., a database containing records identifying
products being manufactured with parameter data for each product,
such as colors and models) or product sales databases (e.g., a
database containing individual data records identifying details of
individual product sales).
[0039] The unstructured data may be presented to the computing
environment 114 in different forms such as a flat file or a
conglomerate of data records and may have data values and
accompanying time stamps. The computing environment 114 may be used
to analyze the unstructured data in a variety of ways to determine
the best way to structure (e.g., hierarchically) that data, such
that the structured data is tailored to a type of further analysis
that a user wishes to perform on the data. For example, after being
processed, the unstructured time stamped data may be aggregated by
time (e.g., into daily time period units) to generate time series
data and/or structured hierarchically according to one or more
dimensions (e.g., parameters, attributes, and/or variables). For
example, data may be stored in a hierarchical data structure, such
as a ROLAP OR MOLAP database, or may be stored in another tabular
form, such as in a flat-hierarchy form.
[0040] Data transmission network 100 may also include one or more
server farms 106. Computing environment 114 may route select
communications or data to the one or more sever farms 106 or one or
more servers within the server farms. Server farms 106 can be
configured to provide information in a predetermined manner. For
example, server farms 106 may access data to transmit in response
to a communication. Server farms 106 may be separately housed from
each other device within data transmission network 100, such as
computing environment 114, and/or may be part of a device or
system.
[0041] Server farms 106 may host a variety of different types of
data processing as part of data transmission network 100. Server
farms 106 may receive a variety of different data from network
devices, from computing environment 114, from cloud network 116, or
from other sources. The data may have been obtained or collected
from one or more sensors, as inputs from a control database, or may
have been received as inputs from an external system or device.
Server farms 106 may assist in processing the data by turning raw
data into processed data based on one or more rules implemented by
the server farms. For example, sensor data may be analyzed to
determine changes in an environment over time or in real-time.
[0042] Data transmission network 100 may also include one or more
cloud networks 116. Cloud network 116 may include a cloud
infrastructure system that provides cloud services. In certain
embodiments, services provided by the cloud network 116 may include
a host of services that are made available to users of the cloud
infrastructure system on demand. Cloud network 116 is shown in FIG.
1 as being connected to computing environment 114 (and therefore
having computing environment 114 as its client or user), but cloud
network 116 may be connected to or utilized by any of the devices
in FIG. 1. Services provided by the cloud network can dynamically
scale to meet the needs of its users. The cloud network 116 may
comprise one or more computers, servers, and/or systems. In some
embodiments, the computers, servers, and/or systems that make up
the cloud network 116 are different from the user's own on-premises
computers, servers, and/or systems. For example, the cloud network
116 may host an application, and a user may, via a communication
network such as the Internet, on demand, order and use the
application.
[0043] While each device, server and system in FIG. 1 is shown as a
single device, it will be appreciated that multiple devices may
instead be used. For example, a set of network devices can be used
to transmit various communications from a single user, or remote
server 140 may include a server stack. As another example, data may
be processed as part of computing environment 114.
[0044] Each communication within data transmission network 100
(e.g., between client devices, between a device and connection
management system 150, between servers 106 and computing
environment 114 or between a server and a device) may occur over
one or more networks 108. Networks 108 may include one or more of a
variety of different types of networks, including a wireless
network, a wired network, or a combination of a wired and wireless
network. Examples of suitable networks include the Internet, a
personal area network, a local area network (LAN), a wide area
network (WAN), or a wireless local area network (WLAN). A wireless
network may include a wireless interface or combination of wireless
interfaces. As an example, a network in the one or more networks
108 may include a short-range communication channel, such as a
Bluetooth or a Bluetooth Low Energy channel. A wired network may
include a wired interface. The wired and/or wireless networks may
be implemented using routers, access points, bridges, gateways, or
the like, to connect devices in the network 108, as will be further
described with respect to FIG. 2. The one or more networks 108 can
be incorporated entirely within or can include an intranet, an
extranet, or a combination thereof. In one embodiment,
communications between two or more systems and/or devices can be
achieved by a secure communications protocol, such as secure
sockets layer (SSL) or transport layer security (TLS). In addition,
data and/or transactional details may be encrypted.
[0045] Some aspects may utilize the Internet of Things (IoT), where
things (e.g., machines, devices, phones, sensors) can be connected
to networks and the data from these things can be collected and
processed within the things and/or external to the things. For
example, the IoT can include sensors in many different devices, and
high value analytics can be applied to identify hidden
relationships and drive increased efficiencies. This can apply to
both big data analytics and real-time analytics. This will be
described further below with respect to FIG. 2.
[0046] As noted, computing environment 114 may include a
communications grid 120 and a transmission network database system
118. Communications grid 120 may be a grid-based computing system
for processing large amounts of data. The transmission network
database system 118 may be for managing, storing, and retrieving
large amounts of data that are distributed to and stored in the one
or more network-attached data stores 110 or other data stores that
reside at different locations within the transmission network
database system 118. The compute nodes in the grid-based computing
system 120 and the transmission network database system 118 may
share the same processor hardware, such as processors that are
located within computing environment 114.
[0047] FIG. 2 illustrates an example network including an example
set of devices communicating with each other over an exchange
system and via a network, according to embodiments of the present
technology. As noted, each communication within data transmission
network 100 may occur over one or more networks. System 200
includes a network device 204 configured to communicate with a
variety of types of client devices, for example client devices 230,
over a variety of types of communication channels.
[0048] As shown in FIG. 2, network device 204 can transmit a
communication over a network (e.g., a cellular network via a base
station 210). The communication can be routed to another network
device, such as network devices 205-209, via base station 210. The
communication can also be routed to computing environment 214 via
base station 210. For example, network device 204 may collect data
either from its surrounding environment or from other network
devices (such as network devices 205-209) and transmit that data to
computing environment 214.
[0049] Although network devices 204-209 are shown in FIG. 2 as a
mobile phone, laptop computer, tablet computer, temperature sensor,
motion sensor, and audio sensor respectively, the network devices
may be or include sensors that are sensitive to detecting aspects
of their environment. For example, the network devices may include
sensors such as water sensors, power sensors, electrical current
sensors, chemical sensors, optical sensors, pressure sensors,
geographic or position sensors (e.g., GPS), velocity sensors,
acceleration sensors, flow rate sensors, among others. Examples of
characteristics that may be sensed include force, torque, load,
strain, position, temperature, air pressure, fluid flow, chemical
properties, resistance, electromagnetic fields, radiation,
irradiance, proximity, acoustics, moisture, distance, speed,
vibrations, acceleration, electrical potential, electrical current,
among others. The sensors may be mounted to various components used
as part of a variety of different types of systems (e.g., a
financial operation). The network devices may detect and record
data related to the environment that it monitors and transmit that
data to computing environment 214.
[0050] In another example, another type of system that may include
various sensors that collect data to be processed and/or
transmitted to a computing environment according to certain
embodiments includes a home automation or similar automated network
in a different environment, such as an office space, school, public
space, sports venue, or a variety of other locations. Network
devices in such an automated network may include network devices
that allow a user to access, control, and/or configure various home
appliances located within the user's home (e.g., a television,
radio, light, fan, humidifier, sensor, microwave, iron, and/or the
like), or outside of the user's home (e.g., exterior motion
sensors, exterior lighting, garage door openers, sprinkler systems,
or the like). For example, network device 202 may include a home
automation switch that may be coupled with a home appliance. In
another embodiment, a network device can allow a user to access,
control, and/or configure devices, such as office-related devices
(e.g., copy machine, printer, or fax machine), audio and/or video
related devices (e.g., a receiver, a speaker, a projector, a DVD
player, or a television), media-playback devices (e.g., a compact
disc player, a CD player, or the like), computing devices (e.g., a
home computer, a laptop computer, a tablet, a personal digital
assistant (PDA), a computing device, or a wearable device),
lighting devices (e.g., a lamp or recessed lighting), devices
associated with a security system, devices associated with an alarm
system, devices that can be operated in an automobile (e.g., radio
devices, navigation devices), and/or the like. Data may be
collected from such various sensors in raw form, or data may be
processed by the sensors to create parameters or other data either
developed by the sensors based on the raw data or assigned to the
system by a client or other controlling device.
[0051] In another example, another type of system that may include
various sensors that collect data to be processed and/or
transmitted to a computing environment according to certain
embodiments includes a power or energy grid. A variety of different
network devices may be included in an energy grid, such as various
devices within one or more power plants, energy farms (e.g., wind
farm, solar farm, among others) energy storage facilities,
factories, homes and businesses of consumers, among others. One or
more of such devices may include one or more sensors that detect
energy gain or loss, electrical input or output or loss, and a
variety of other efficiencies. These sensors may collect data to
inform users of how the energy grid, and individual devices within
the grid, may be functioning and how they may be made more
efficient.
[0052] Network device sensors may also perform processing on data
it collects before transmitting the data to the computing
environment 214, or before deciding whether to transmit data to the
computing environment 214. For example, network devices may
determine whether data collected meets certain rules, for example
by comparing data or values computed from the data and comparing
that data to one or more thresholds. The network device may use
this data and/or comparisons to determine if the data should be
transmitted to the computing environment 214 for further use or
processing.
[0053] Computing environment 214 may include machines 220 and 240.
Although computing environment 214 is shown in FIG. 2 as having two
machines, 220 and 240, computing environment 214 may have only one
machine or may have more than two machines. The machines that make
up computing environment 214 may include specialized computers,
servers, or other machines that are configured to individually
and/or collectively process large amounts of data. The computing
environment 214 may also include storage devices that include one
or more databases of structured data, such as data organized in one
or more hierarchies, or unstructured data. The databases may
communicate with the processing devices within computing
environment 214 to distribute data to them. Since network devices
may transmit data to computing environment 214, that data may be
received by the computing environment 214 and subsequently stored
within those storage devices. Data used by computing environment
214 may also be stored in data stores 235, which may also be a part
of or connected to computing environment 214.
[0054] Computing environment 214 can communicate with various
devices via one or more routers 225 or other inter-network or
intra-network connection components. For example, computing
environment 214 may communicate with devices 230 via one or more
routers 225. Computing environment 214 may collect, analyze and/or
store data from or pertaining to communications, client device
operations, client rules, and/or user-associated actions stored at
one or more data stores 235. Such data may influence communication
routing to the devices within computing environment 214, how data
is stored or processed within computing environment 214, among
other actions.
[0055] Notably, various other devices can further be used to
influence communication routing and/or processing between devices
within computing environment 214 and with devices outside of
computing environment 214. For example, as shown in FIG. 2,
computing environment 214 may include a machine 240, such as a web
server. Thus, computing environment 214 can retrieve data of
interest, such as client information (e.g., product information,
client rules, etc.), technical product details, news, current or
predicted weather, and so on.
[0056] In addition to computing environment 214 collecting data
(e.g., as received from network devices, such as sensors, and
client devices or other sources) to be processed as part of a big
data analytics project, it may also receive data in real time as
part of a streaming analytics environment. As noted, data may be
collected using a variety of sources as communicated via different
kinds of networks or locally. Such data may be received on a
real-time streaming basis. For example, network devices may receive
data periodically from network device sensors as the sensors
continuously sense, monitor and track changes in their
environments. Devices within computing environment 214 may also
perform pre-analysis on data it receives to determine if the data
received should be processed as part of an ongoing project. The
data received and collected by computing environment 214, no matter
what the source or method or timing of receipt, may be processed
over a period of time for a client to determine results data based
on the client's needs and rules.
[0057] FIG. 3 illustrates a conceptual overview of a system 300
that may generally be representative of a distributed cloud-based
computing system or another type of computing network in that one
or more techniques described herein may be implemented according to
various embodiments. As shown in FIG. 3, system 300 may generally
include computing resources (CRs) 302-y, where y may be any
positive integer, to compute information and data. The computing
resources 302 may include resources of multiple types, such as--for
example--processors, co-processors, fully-programmable gate arrays
(FPGAs), memory, networking equipment, circuit boards, storage, and
other computing equipment. The embodiments are not limited to these
examples.
[0058] The computing resources 302 may be included as part of a
computer, such as a server, server farm, blade server, a server
sled, or any other type of server or computing device, and may be
within one or more racks 304. In embodiments, the racks 304 may be
part of one or more data centers 308 and may be coupled with each
other via various networking equipment. For example, the racks 304
within a data center 308 may be coupled with each via a fabric 303.
The fabric 303 may include a combination of electrical and/or
optical signaling media, and high bandwidth interconnects, such as
Gigabit Ethernet, 10 Gigabit Ethernet, 100 Gigabit Ethernet,
InfiniB and, Peripheral Component Interconnect (PCI) Express
(PCIe), and so forth. Further, the fabric 303 may include switching
infrastructure, such as switches, routers, gateways, and so forth.
The fabric 303 is configured such that any rack 304 may send
signals to (and receive signals from) each other racks 304 within a
data center 308 to communicate data and information. In
embodiments, the fabric 303 may be coupled with networking
infrastructure 305 such that it enables communication of signals
between racks of one data center 308 with racks 304 of another data
center 308 to communicate data and information. For example, FIG. 3
illustrates racks 304 of data center 308-1 coupled with racks 304
of data center 308-2 via fabrics 303, networking infrastructure
305, and the cloud-based infrastructure 307.
[0059] In embodiments, the networking infrastructure 305 includes
networking equipment, such as routers, firewalls, switches,
gateways, cabling, and so forth to communicate data and information
between a data center 308 and with the cloud-based infrastructure
307 and another data center 308. For example, the networking
infrastructure 305 may include edge access routers, edge access
switches, and edge firewalls capable of communicating with core
routers, core switches, and core firewalls of the cloud-based
infrastructure 307. The core networking equipment of the
cloud-based infrastructure 307 may couple with edge networking
equipment of another data center 308 to enable communication
between data centers 308. Note that embodiments are not limited in
this manner, and the networking infrastructure 305 and/or
cloud-based infrastructure 307 may include other networking
equipment, servers, relays, interconnects, and so forth to enable
communication between a data center 308 and other data centers
308.
[0060] In one example, the system 300 may be a distributed
cloud-based computing system to provide a financial service
platform. The system 300 may process data and information, such as
financial data and financial information, to provide financial
services, for example. The financial services include, but are not
limited to, investment and asset management services, active equity
management services, active quantitative equity services, cash fund
services, alternatives services, currency management services,
index investing services, electronic trading services, multi-asset
services, investment research services, investment trading
services, accounting services, custody services, fund
administration services, outsourcing services, performance
measurement services, portfolio analysis services, data analytics
services, investment analytics services,
benchmark/indices/indicator services, D-as-a-Service (DaaS)
services, and so forth. Embodiments are not limited to these
examples. To provide these financial services, the system 300 may
dynamically pool or compose a plurality of the computing resources
302 together within a data center 308 and/or among data centers 308
in the cloud via the cloud-based infrastructure 307. In one
example, computing resources 302 of data center 308-1 may be
composed with computing resources 302 of data center 308-2 to
process data and information, e.g., a job, to provide a financial
service. Once the job complete, the system 300 may decompose the
composed computing resources 302 and make the computing resources
302 available to process another job. Note that in embodiments, the
system 300 may enable multiple instances of pooled or composed
computing resources 302 to provide data and information processing
in parallel and embodiments are not limited in this manner.
[0061] In embodiments, system 300 may be coupled with one or more
other systems, such as investment trading systems, banking systems,
regulatory systems, risk management systems, performance systems,
accounting system, data warehouse systems, financial institution
system, and so forth. These other systems may be coupled with
system 300 via networking, such as the networking infrastructure
305 and the cloud-based infrastructure. Embodiments are not limited
in this manner.
[0062] FIG. 4 depicts a block diagram 400 describing exemplary
logic and data flows through a computing architecture. Data 402 may
be received in a data storage 404. The data 402 may include
discrete units of data and/or one or more data streams (e.g.,
communication channels that repeatedly provide one or more data
units at a given rate). The data 402 may include, for example, data
relating to an individual user (e.g., a financial client),
aggregate data (e.g., reflecting conditions in a market, such as a
financial market), cancelations of previously-received data,
corrections of previously-received data, etc. In some cases,
cancelations and/or corrections may be received that cancels or
corrects other data items that have not yet been received (e.g.,
due to the order in which the data was transmitted or batched,
network conditions, data losses, etc.).
[0063] The data storage 404 may include hardware, software, or a
combination of hardware and software suitable for storing data. The
data storage 404 may include one or more data structures, such as a
log 406 or a database. The data structures may be configured to
store and organize the data, and/or to facilitate retrieval of the
data. In some embodiments, the log 406 may be organized
chronologically (e.g., in the order in which the data 402 was
received by the data storage 404, in a time-stamp order of the data
402, etc.). In some embodiments, the log 406 is a persistent and/or
immutable log which allows individual data records to be written,
but not to be directly deleted or changed.
[0064] In some embodiments, the data 402 stored in the data storage
404 may be subjected to one or more filters 408. The filters 408
may include data governance filters which, for example, match one
or more rules against the data 402 and selectively pass the data
402 to other components in the architecture.
[0065] The data storage 404 and/or the filter(s) 408 may provide
information to a machine learning model 410, such as an artificial
neural network (ANN). The underlying model 410 may be configured to
learn associations from patterns in the data 402, to predict future
trends based on historical data observations, and to provide
insights into why the data 402 appears the way that it does.
[0066] A library 414 of microservices 412-i may make use the data
(e.g., the raw data stored in the data storage 404, the filtered
data as presented by the filters 408, information output from the
machine learning model 410, or various combinations of these types
of data). Each microservice 412-i may represent an atomic computing
unit configured to perform a defined task (e.g., computing a value
for a financial variable for certain subsets of the data 402). The
microservices 412-i may be used individually, or variously combined
into macroservices 416-i. The macroservices 416-i may represent
more complex operations in which the outputs of various
microservices 412-i are combined or otherwise used to perform a
specified task.
[0067] For instance, one macroservice 416-1 may use the outputs of
various microservices 412-i to generate a report 418 (such as a
financial report, disclosure form, etc.). In another example, an
entity (such as a financial regulator) may issue a request 420 via
a macroservice 416-2, and the microservice 416-2 may perform
various operations to comply with the request (e.g., calling on
another macroservice 416-1 to generate a report responsive to the
request 420, correcting data 402 in the data storage 404, etc.). In
some embodiments, macroservices 416-i may themselves be combined
together to form other macroservices 416-i.
[0068] The microservices 412-i and/or the macroservices 416-i may
be exposed to a third party (e.g., by use of an application
programming interface, or "API"). In some cases, a single entity
may provide the microservices 412-i and the macroservices 416-i. In
other cases, one entity may provide the library 414 of
microservices 412-i, and another entity may use the microservices
414 to generate their own customized macroservices 416-i.
[0069] The machine learning model 410 may be generated and/or
refined via a machine learning process, such as the one depicted in
the flow chart of FIG. 4. Machine learning is a branch of
artificial intelligence that relates to mathematical models that
can learn from, categorize, and make predictions about data. Such
mathematical models, which can be referred to as machine-learning
models, can classify input data among two or more classes; cluster
input data among two or more groups; predict a result based on
input data; identify patterns or trends in input data; identify a
distribution of input data in a space; or any combination of these.
Examples of machine-learning models can include (i) neural
networks; (ii) decision trees, such as classification trees and
regression trees; (iii) classifiers, such as Naive bias
classifiers, logistic regression classifiers, ridge regression
classifiers, random forest classifiers, least absolute shrinkage
and selector (LASSO) classifiers, and support vector machines; (iv)
clusterers, such as k-means clusterers, mean-shift clusterers, and
spectral clusterers; (v) factorizers, such as factorization
machines, principal component analyzers and kernel principal
component analyzers; and (vi) ensembles or other combinations of
machine-learning models. In some examples, neural networks can
include deep neural networks, feed-forward neural networks,
recurrent neural networks, convolutional neural networks, radial
basis function (RBF) neural networks, echo state neural networks,
long short-term memory neural networks, bi-directional recurrent
neural networks, gated neural networks, hierarchical recurrent
neural networks, stochastic neural networks, modular neural
networks, spiking neural networks, dynamic neural networks,
cascading neural networks, neuro-fuzzy neural networks, or any
combination of these.
[0070] Different machine-learning models may be used
interchangeably to perform a task. Examples of tasks that can be
performed at least partially using machine-learning models include
various types of scoring; bioinformatics; cheminformatics; software
engineering; fraud detection; customer segmentation; generating
online recommendations; adaptive websites; determining customer
lifetime value; search engines; placing advertisements in real time
or near real time; classifying DNA sequences; affective computing;
performing natural language processing and understanding; object
recognition and computer vision; robotic locomotion; playing games;
optimization and metaheuristics; detecting network intrusions;
medical diagnosis and monitoring; or predicting when an asset, such
as a machine, will need maintenance.
[0071] Machine-learning models can be constructed through an at
least partially automated (e.g., with little or no human
involvement) process called training. During training, input data
can be iteratively supplied to a machine-learning model to enable
the machine-learning model to identify patterns related to the
input data or to identify relationships between the input data and
output data. With training, the machine-learning model can be
transformed from an untrained state to a trained state. Input data
can be split into one or more training sets and one or more
validation sets, and the training process may be repeated multiple
times. The splitting may follow a k-fold cross-validation rule, a
leave-one-out-rule, a leave-p-out rule, or a holdout rule. An
overview of training and using a machine-learning model is
described below with respect to the flow chart 500 of FIG. 5.
[0072] In block 502, training data is received. In some examples,
the training data is received from a remote database or a local
database, constructed from various subsets of data, or input by a
user. The training data can be used in its raw form for training a
machine-learning model or pre-processed into another form, which
can then be used for training the machine-learning model. For
example, the raw form of the training data can be smoothed,
truncated, aggregated, clustered, or otherwise manipulated into
another form, which can then be used for training the
machine-learning model.
[0073] In block 504, a machine-learning model is trained using the
training data. The machine-learning model can be trained in a
supervised, unsupervised, or semi-supervised manner. In supervised
training, each input in the training data is correlated to a
desired output. This desired output may be a scalar, a vector, or a
different type of data structure such as text or an image. This may
enable the machine-learning model to learn a mapping between the
inputs and desired outputs. In unsupervised training, the training
data includes inputs, but not desired outputs, so that the
machine-learning model has to find structure in the inputs on its
own. In semi-supervised training, only some of the inputs in the
training data are correlated to desired outputs.
[0074] In block 506, the machine-learning model is evaluated. For
example, an evaluation dataset can be obtained, for example, via
user input or from a database. The evaluation dataset can include
inputs correlated to desired outputs. The inputs can be provided to
the machine-learning model and the outputs from the
machine-learning model can be compared to the desired outputs. If
the outputs from the machine-learning model closely correspond with
the desired outputs, the machine-learning model may have a high
degree of accuracy. For example, if 90% or more of the outputs from
the machine-learning model are the same as the desired outputs in
the evaluation dataset, the machine-learning model may have a high
degree of accuracy. Otherwise, the machine-learning model may have
a low degree of accuracy. The 90% value is an example only. A
realistic and desirable accuracy percentage is dependent on the
problem and the data.
[0075] In some examples, if the machine-learning model has an
inadequate degree of accuracy for a particular task, the process
can return to block 504, where the machine-learning model can be
further trained using additional training data or otherwise
modified to improve accuracy. If the machine-learning model has an
adequate degree of accuracy for the particular task, the process
can continue to block 508.
[0076] In block 508, new data is received. In some examples, the
new data is received from a remote database or a local database,
constructed from various subsets of data, or input by a user. The
new data may be unknown to the machine-learning model. For example,
the machine-learning model may not have previously processed or
analyzed the new data.
[0077] In block 510, the trained machine-learning model is used to
analyze the new data and provide a result. For example, the new
data can be provided as input to the trained machine-learning
model. The trained machine-learning model can analyze the new data
and provide a result that includes a classification of the new data
into a particular class, a clustering of the new data into a
particular group, a prediction based on the new data, or any
combination of these.
[0078] In block 512, the result is post-processed. For example, the
result can be added to, multiplied with, or otherwise combined with
other data as part of a job. As another example, the result can be
transformed from a first format, such as a time series format, into
another format, such as a count series format. Any number and
combination of operations can be performed on the result during
post-processing.
[0079] A more specific example of a machine-learning model is the
neural network 600 shown in FIG. 6. The neural network 600 is
represented as multiple layers of interconnected neurons, such as
neuron 608, that can exchange data between one another. The layers
include an input layer 602 for receiving input data, a hidden layer
604, and an output layer 606 for providing a result. The hidden
layer 604 is referred to as hidden because it may not be directly
observable or have its input directly accessible during the normal
functioning of the neural network 600. Although the neural network
600 is shown as having a specific number of layers and neurons for
exemplary purposes, the neural network 600 can have any number and
combination of layers, and each layer can have any number and
combination of neurons.
[0080] The neurons and connections between the neurons can have
numeric weights, which can be tuned during training. For example,
training data can be provided to the input layer 602 of the neural
network 600, and the neural network 600 can use the training data
to tune one or more numeric weights of the neural network 600.
[0081] In some examples, the neural network 600 can be trained
using backpropagation. Backpropagation can include determining a
gradient of a particular numeric weight based on a difference
between an actual output of the neural network 600 and a desired
output of the neural network 600. Based on the gradient, one or
more numeric weights of the neural network 600 can be updated to
reduce the difference, thereby increasing the accuracy of the
neural network 600. This process can be repeated multiple times to
train the neural network 600. For example, this process can be
repeated hundreds or thousands of times to train the neural network
600.
[0082] In some examples, the neural network 600 is a feed-forward
neural network. In a feed-forward neural network, every neuron only
propagates an output value to a subsequent layer of the neural
network 600. For example, data may only move one direction
(forward) from one neuron to the next neuron in a feed-forward
neural network.
[0083] In other examples, the neural network 600 is a recurrent
neural network. A recurrent neural network can include one or more
feedback loops, allowing data to propagate in both forward and
backward through the neural network 600. This can allow for
information to persist within the recurrent neural network. For
example, a recurrent neural network can determine an output based
at least partially on information that the recurrent neural network
has seen before, giving the recurrent neural network the ability to
use previous input to inform the output.
[0084] In some examples, the neural network 600 operates by
receiving a vector of numbers from one layer; transforming the
vector of numbers into a new vector of numbers using a matrix of
numeric weights, a nonlinearity, or both; and providing the new
vector of numbers to a subsequent layer of the neural network 600.
Each subsequent layer of the neural network 600 can repeat this
process until the neural network 600 outputs a final result at the
output layer 606. For example, the neural network 600 can receive a
vector of numbers as an input at the input layer 602. The neural
network 600 can multiply the vector of numbers by a matrix of
numeric weights to determine a weighted vector. The matrix of
numeric weights can be tuned during the training of the neural
network 600. The neural network 600 can transform the weighted
vector using a nonlinearity, such as a sigmoid tangent or the
hyperbolic tangent. In some examples, the nonlinearity can include
a rectified linear unit, which can be expressed using the following
equation:
y=max(x,0)
where y is the output and x is an input value from the weighted
vector. The transformed output can be supplied to a subsequent
layer, such as the hidden layer 604, of the neural network 600. The
subsequent layer of the neural network 600 can receive the
transformed output, multiply the transformed output by a matrix of
numeric weights and a nonlinearity, and provide the result to yet
another layer of the neural network 600. This process continues
until the neural network 600 outputs a final result at the output
layer 606.
[0085] Other examples of the present disclosure may include any
number and combination of machine-learning models having any number
and combination of characteristics. The machine-learning model(s)
can be trained in a supervised, semi-supervised, or unsupervised
manner, or any combination of these. The machine-learning model(s)
can be implemented using a single computing device or multiple
computing devices, such as the communications system discussed
herein.
[0086] Implementing some examples of the present disclosure at
least in part by using machine-learning models can reduce the total
number of processing iterations, time, memory, electrical power, or
any combination of these consumed by a computing device when
analyzing data. For example, a neural network may more readily
identify patterns in data than other approaches. This may enable
the neural network to analyze the data using fewer processing
cycles and less memory than other approaches, while obtaining a
similar or greater level of accuracy.
[0087] The methods, systems, and functionality described herein may
be embodied as instructions on a computer readable medium or as
part of a computing architecture. FIG. 7 illustrates an embodiment
of an exemplary computing architecture 700 suitable for
implementing various embodiments described herein. In one
embodiment, the computing architecture 700 may comprise or be
implemented as part of an electronic device, such as a computer
701. The embodiments are not limited in this context.
[0088] As used in this application, the terms "system" and
"component" are intended to refer to a computer-related entity,
either hardware, a combination of hardware and software, software,
or software in execution, examples of which are provided by the
exemplary computing architecture 700. For example, a component can
be, but is not limited to being, a process executing on a
processor, a processor, a hard disk drive, multiple storage drives
(of optical and/or magnetic storage medium), an object, an
executable, a thread of execution, a program, and/or a computer. By
way of illustration, both an application running on a server and
the server can be a component. One or more components can reside
within a process and/or thread of execution, and a component can be
localized on one computer and/or distributed between two or more
computers. Further, components may be communicatively coupled to
each other by various types of communications media to coordinate
operations. The coordination may involve the uni-directional or
bi-directional exchange of information. For instance, the
components may communicate information in the form of signals
communicated over the communications media.
[0089] The information can be implemented as signals allocated to
various signal lines. In such allocations, each message is a
signal. Further embodiments, however, may alternatively employ data
messages. Such data messages may be sent across various
connections. Exemplary connections include parallel interfaces,
serial interfaces, and bus interfaces.
[0090] The computing architecture 700 includes various common
computing elements, such as one or more processors, multi-core
processors, co-processors, memory units, chipsets, controllers,
peripherals, interfaces, oscillators, timing devices, video cards,
audio cards, multimedia input/output (I/O) components, power
supplies, and so forth. The embodiments, however, are not limited
to implementation by the computing architecture 700.
[0091] As shown in FIG. 7, the computing architecture 700 comprises
a processing unit 702, a system memory 704 and a chipset 706. The
processing unit 702 can be any of various commercially available
processors, including without limitation an AMD.RTM. Athlon.RTM.,
Duron.RTM. and Opteron.RTM. processors; ARM.RTM. application,
embedded and secure processors; IBM.RTM. and Motorola.RTM.
DragonBall.RTM. and PowerPC.RTM. processors; IBM and Sony.RTM. Cell
processors; Intel.RTM. Celeron.RTM., Core (2) Duo.RTM., Core
i9.TM., Core m3.TM., vPro.TM., Itanium.RTM., Pentium.RTM.,
Xeon.RTM., and XScale.RTM. processors; and similar processors. Dual
microprocessors, multi-core processors, and other multi-processor
architectures may also be employed as the processing unit 702.
[0092] In some embodiments, the processing unit 702 couples with
the chipset 706 via a highspeed serial link 703 and couples with
the system memory 704 via a highspeed serial link 705. In other
embodiments, the processing unit 702 may couple with the chipset
706 and possibly other processor units via a system bus and may
couple with the system memory 704 via the chipset 706. In further
embodiments, the processing unit 702 and the chipset may reside in
a System-On-Chip (SoC) package.
[0093] The chipset 706 provides an interface for system components
including, but not limited to, the system memory 704 to the
processing unit 702. The chipset 706 may couple with any of several
types of bus structure that may further interconnect to a memory
bus (with or without a memory controller), a peripheral bus, and a
local bus using any of a variety of commercially available bus
architectures. Interface adapters 722, 724, 726, 728, 740, 752,
etc., may connect to the chipset 706 via a slot architecture.
Example slot architectures may include without limitation
Accelerated Graphics Port (AGP), Card Bus, (Extended) Industry
Standard Architecture ((E)ISA), Micro Channel Architecture (MCA),
NuBus, Peripheral Component Interconnect (Extended) (PCI(X)), PCI
Express, Personal Computer Memory Card International Association
(PCMCIA), and the like.
[0094] The computing architecture 700 may comprise or implement
various articles of manufacture. An article of manufacture may
comprise a computer-readable storage medium to store logic.
Examples of a computer-readable storage medium may include any
tangible media capable of storing electronic data, including
volatile memory or non-volatile memory, removable or non-removable
memory, erasable or non-erasable memory, writeable or re-writeable
memory, and so forth. Examples of logic may include executable
computer program instructions implemented using any suitable type
of code, such as source code, compiled code, interpreted code,
executable code, static code, dynamic code, object-oriented code,
visual code, and the like. Embodiments may also be at least partly
implemented as instructions contained in or on a non-transitory
computer-readable medium, which may be read and executed by one or
more processors to enable performance of the operations described
herein.
[0095] The system memory 704 may include various types of
computer-readable storage media in the form of one or more higher
speed memory units, such as read-only memory (ROM), random-access
memory (RAM), dynamic RAM (DRAM), Double-Data-Rate DRAM (DDRAM),
synchronous DRAM (SDRAM), static RAM (SRAM), programmable ROM
(PROM), erasable programmable ROM (EPROM), electrically erasable
programmable ROM (EEPROM), flash memory, polymer memory such as
ferroelectric polymer memory, ovonic memory, phase change or
ferroelectric memory, silicon-oxide-nitride-oxide-silicon (SONOS)
memory, magnetic or optical cards, an array of devices such as
Redundant Array of Independent Disks (RAID) drives, solid state
memory devices (e.g., USB memory, solid state drives (SSD) and any
other type of storage media suitable for storing information. In
the illustrated embodiment shown in FIG. 7, the system memory 704
can include non-volatile memory 708 and/or volatile memory 710. A
basic input/output system (BIOS) can be stored in the non-volatile
memory 708.
[0096] The computing architecture 700 may include various types of
computer-readable storage media in the form of one or more lower
speed memory units, including an internal (or external) hard disk
drive (HDD) 712, a magnetic floppy disk drive (FDD) 714 to read
from or write to a removable magnetic disk 716, and an optical disk
drive 718 to read from or write to a removable optical disk 720
(e.g., a CD-ROM or DVD). The HDD 712, FDD 714 and optical disk
drive 720 can be connected to the system bus 706 by an HDD
interface 722, an FDD interface 724 and an optical drive interface
726, respectively. The HDD interface 722 for external drive
implementations can include at least one or both of Universal
Serial Bus (USB) and IEEE 1394 interface technologies.
[0097] The drives and associated computer-readable media provide
volatile and/or nonvolatile storage of data, data structures,
computer-executable instructions, and so forth. For example, a
number of program modules can be stored in the drives and memory
units 708, 712, including an operating system 728, one or more
application programs 730, other program modules 732, and program
data 734. In one embodiment, the one or more application programs
730, other program modules 732, and program data 734 can include,
for example, the various applications and/or components described
herein.
[0098] A user may enter commands and information into the computer
701 through one or more wire/wireless input devices, for example, a
keyboard 736 and a pointing device, such as a mouse 738. Other
input devices may include microphones, infra-red (IR) remote
controls, radio-frequency (RF) remote controls, game pads, stylus
pens, card readers, dongles, finger print readers, gloves, graphics
tablets, joysticks, keyboards, retina readers, touch screens (e.g.,
capacitive, resistive, etc.), trackballs, trackpads, sensors,
styluses, and the like. These and other input devices are often
connected to the processing unit 702 through an input device
interface 740 that is coupled to the chipset 706, but can be
connected by other interfaces such as a parallel port, IEEE 694
serial port, a game port, a USB port, an IR interface, and so
forth.
[0099] A monitor 742 or other type of display device is also
connected to the chip set 706 via an interface, such as a video
adaptor 728. The monitor 742 may be internal or external to the
computer 701. In addition to the monitor 742, a computer typically
includes other peripheral output devices, such as speakers,
printers, and so forth.
[0100] The computer 701 may operate in a networked environment
using logical connections via wire and/or wireless communications
to one or more remote computers, such as a remote computer 744. The
remote computer 744 can be a workstation, a server computer, a
router, a personal computer, portable computer,
microprocessor-based entertainment appliance, a peer device or
other common network node, and typically includes many of or all
the elements described relative to the computer 701, although, for
purposes of brevity, only a memory/storage device 746 is
illustrated. The logical connections depicted include wire/wireless
connectivity to a local area network (LAN) 748 and/or larger
networks, for example, a wide area network (WAN) 750. Such LAN and
WAN networking environments are commonplace in offices and
companies, and facilitate enterprise-wide computer networks, such
as intranets, all which may connect to a global communications
network, for example, the Internet.
[0101] When used in a LAN networking environment, the computer 701
is connected to the LAN 748 through a wire and/or wireless
communication network interface or adaptor 752. The adaptor 752 can
facilitate wire and/or wireless communications to the LAN 748,
which may also include a wireless access point disposed thereon for
communicating with the wireless functionality of the adaptor
752.
[0102] When used in a WAN networking environment, the computer 701
can include a modem 754, or is connected to a communications server
on the WAN 750, or has other means for establishing communications
over the WAN 750, such as by way of the Internet. The modem 754,
which can be internal or external and a wire and/or wireless
device, connects to the chipset 706 via the input device interface
740. In a networked environment, program modules depicted relative
to the computer 701, or portions thereof, can be stored in the
remote memory/storage device 746. It will be appreciated that the
network connections shown are exemplary and other means of
establishing a communications link between the computers can be
used.
[0103] The computer 701 is operable to communicate with wire and
wireless devices or entities using the IEEE 802 family of
standards, such as wireless devices operatively disposed in
wireless communication (e.g., IEEE 802.13 over-the-air modulation
techniques). This includes at least Wi-Fi (or Wireless Fidelity),
WiMax, and Bluetooth.TM. wireless technologies, among others. Thus,
the communication can be a predefined structure as with a
conventional network or simply an ad hoc communication between at
least two devices. Wi-Fi networks use radio technologies called
IEEE 802.13x (a, b, g, n, etc.) to provide secure, reliable, fast
wireless connectivity. A Wi-Fi network can be used to connect
computers to each other, to the Internet, and to wire networks
(which use IEEE 802.3-related media and functions).
[0104] Some aspects may utilize the Internet of Things (IoT), where
things (e.g., machines, devices, phones, sensors) can be connected
to networks and the data from these things can be collected and
processed within the things and/or external to the things. For
example, with the IoT, sensors may be deployed in many different
devices, and high-value analytics can be applied to identify hidden
relationships and drive increased efficiencies. This can apply to
both Big Data analytics and real-time (streaming) analytics.
[0105] Some systems may use Hadoop.RTM., an open-source framework
for storing and analyzing big data in a distributed computing
environment. Apache.TM. Hadoop.RTM. is an open-source software
framework for distributed computing. For example, some grid systems
may be implemented as a multi-node Hadoop.RTM. cluster, as
understood by a person of skill in the art. Some systems may use
cloud computing, which can enable ubiquitous, convenient, on-demand
network access to a shared pool of configurable computing resources
(e.g., networks, servers, storage, applications and services) that
can be rapidly provisioned and released with minimal management
effort or service provider interaction.
[0106] FIG. 8 illustrates an example of an operating environment
800 that may be representative of various embodiments. The
operating environment 800 depicted in FIG. 8 may implement
computational model processes according to various embodiments. In
some embodiments, operating environment 800 may include or
implement a computational model process that includes a plurality
of workflows, including, for example a management workflow, a data
workflow, and/or a deployment workflow. In some embodiments, a
computational model process according to some embodiments may
general flow in the order of paths A-J.
[0107] As shown in FIG. 8, a computational model process may
include requirement and data analysis 802. In exemplary
embodiments, requirement and data analysis 802 may include various
processes to determine the requirements and/or objectives of a
computational model, for example, operative to solve a problem,
perform a function, and/or the like ("model function"). In various
embodiments, requirement and data analysis 802 may include
ingesting, persisting, and/or otherwise obtaining data for
modelling purposes (for instance, "function data" associated with
the function, problem, objective, and/or the like that may be used
to generate a computational model). For instance, requirement and
data analysis 802 may include determining a problem (for instance,
a defined business problem) and/or an objective for a computational
model of interest (for instance, a model function to address the
problem, achieve the objective, and/or the like). For example, a
problem may include recognizing signatures on faxed documents and
an objective may include determining signatures with an accuracy
over a threshold amount. In this example, data analysis may include
accessing relevant documents (function data) with known signatures.
Accordingly, requirement and data analysis 802 may generate model
objectives and model data for computational model process.
[0108] The model objectives and data may be provided to a data
repository platform 818. In some embodiments, data repository
platform 818 may include a data streaming platform operative to,
among other things, publish and subscribe to streams of records,
store streams of records in a fault-tolerant durable way, and/or
process streams of records as they occur. A non-limiting example of
data repository platform may include the Apache Kafka.RTM.
distributed streaming platform. Although Apache Kafka.RTM. platform
may be used as an example data repository platform, embodiments are
not so limited, as any type of data repository platform capable of
operating according to some embodiments is contemplated herein. In
various embodiments, at least a portion of the computational
models, model objectives, and/or model data may be stored in a
repository 820 via data repository platform 818.
[0109] Computational model process may proceed to model building,
tuning, and evaluation 804 ("model building"). Model building 804
may be implemented using various automated machine learning (ML)
platforms including, without limitation, MLflow (provided by
Databricks, Inc. of San Francisco, Calif.) and/or DataRobot
(provided by DataRobot, Inc. of Boston, Mass.). Although MLflow and
DataRobot are used as examples, embodiments are not so limited, as
any automated ML platform capable of operating according to some
embodiments is contemplated herein. In various embodiments, model
building 804 may access models and/or model data via data
repository platform 818 and, in addition, models and associated
information resulting from model building 804 may be stored via
data repository platform 818.
[0110] In some embodiments, computational model processes may
include a champion-challenger process. In some embodiments,
challenger models may be selected during model building 804. In
general, a current computational model being used for a particular
task may be considered a "champion" model. A challenger model may
include a model using a different strategy, process, and/or the
like than the champion model. A challenger model may be tested
using various metrics to determine if performance is superior to a
champion model.
[0111] In various embodiments, once challenger models are selected
during model building 804, a challenger model process 822 may
select challenger models to be provided to a continuous integration
(CI)/continuous delivery (CD) process or pipeline 824 for
deployment in a quality assurance 806 process. A model assessment
process 808 may operate to assess the challenger models 810, for
example, compared with the performance of champion models 812,
using various metrics, including, without limitation, performance
metrics, industrial metrics, user metrics, and/or the like (see,
for example, FIG. 9). In various embodiments, user metrics 948 may
be associated with performance for a particular user, user
environment, files/documents, and/or the like. For example, user
metrics 948 may be associated with whether a model working better
for a user's particular documents, environment, and/or the like.
Accordingly, models evaluated using user metrics 948 may provide
more precise evaluations for a specific user, class of information
(or documents), business case (or business control), and/or the
like.
[0112] In exemplary embodiments, selected models, such as
challenger models 810 that have been selected to be current
champion models 812, may undergo an approval workflow process 814
before proceeding to a model online process 816 operative to
prepare a model to be active. In various embodiments, model
building process 804 may be informed of the current champion model,
for example, when a new challenger model is determined. In various
embodiments, if a challenger model performs better for a task,
process, business case, and/or the like, the challenger model may
be deployed according to some embodiments; otherwise the champion
model may remain in place. The determination of performance of
challenger and champion models may be based on various criteria,
such as time, accuracy, resource requirements, combinations
thereof, and/or the like.
[0113] FIG. 9 illustrates an example of an operating environment
900 that may be representative of various embodiments. The
operating environment 900 depicted in FIG. 9 may implement
computational model processes according to various embodiments. As
shown in FIG. 9, a computational model process may include
determining problem objective 902 and accessing relevant data 904
(such as business data) and providing the corresponding information
to a feature engineering process 906. Relevant ML algorithms 908
may be determined and provided to a ML model building process 910
along with the results of feature engineering process 906. In some
embodiments, output from a sequence of ML algorithm selection 912,
ML model training 914, and model evaluation 916 may be provided to
a candidate model selection process 918 to determine a candidate
model 920.
[0114] Candidate model 920 may be provided to a model serialization
and persistence process 918 and stored in storage 932 and/or
provided to a model registry process 934. In some embodiments,
model registry process 934 may provide candidate model 920, for
example, as processed by model serialization and persistence
process 918, to a model inventory 936. In various embodiments,
model registry process 934 may provide candidate model 920, for
example, as processed by model serialization and persistence
process 918, to a model assessment 905 process. In various
embodiments, model assessment process 905 may operate to evaluate
challenger models against champion models. For example, model
assessment process 905 may determine challenger predictions
(pre-production) 940 for a model problem versus champion
(production or in-production) models 942 using scoring evaluation
data 938, performance metrics 944, industrial metrics 946, and/or
user metrics 948. In some embodiments, user metrics 948 may be or
may include application-specific scoring API 980. In various
embodiments, models and evaluation data (for instance, storage 932,
models inventory 936, and/or scoring evaluation data 938) may be
co-located within causeway 982.
[0115] In various embodiments, model assessment process 905 may
launch a model decay process (for instance, steps 950, 952, 94,
956, 958, 960, 968, 970, 972, and/or 972) operative to determine
whether a model has decayed 950 or otherwise experienced
deteriorating performance. If the model has not experienced model
decay, the model decay evaluation process ends 952. If the model
has experienced decay, a notification for corrective action 954 may
be generated and/or the model may be subjected to model deployment
CI/CD 956, approve workflow 958, and/or release management.
Following model deployment CI/CD 956, a model may be processed via
model initialization 968, mapping with corresponding cognitive
service 970, model go live 972, and/or model monitoring (decay
management) 974.
[0116] FIG. 10 illustrates an example of an operating environment
1000 that may be representative of various embodiments. The
operating environment 1000 depicted in FIG. 10 may implement
computational model processes according to various embodiments. As
shown in FIG. 10, operating environment 1000 may include a model
management framework 1002, micro-services management 1004, and/or
infrastructure 1006.
[0117] Model management framework 1002 may include ML algorithms
1008, data 1012, and/or engineered features 1014. In some
embodiments, ML algorithms 1008, data 1012, and/or engineered
features 1014 may loop 1010 on various features, for example,
selecting a proposed best option for one or more thereof. In
various embodiments, ML algorithms 1008, data 1012, and/or
engineered features 1014 may be used to determine candidate models
1016 via a model test/evaluation process 1018. Models 1020 may
undergo continuous training and/or evaluation 1028. In some
embodiments, models 1020 may be proved to a model registry and
inventory 1022 and/or model monitory 1024. In some embodiments,
models 1020 may be associated with various model objects 1040
including, without limitation model files, model metadata (or
"meta"), versions, parameters, hyper parameters, algorithm
information, data, features, performance matrices, and/or the
like.
[0118] In some embodiments, a computational model process may
include model to service mapping 1026. For example, services 1040
may be associated with or accessed by management micro services
1052 and models 1020 may be mapped, associated with, or otherwise
accessed by cognitive micro services 1054 (such as AI or ML
services). Management micro services 1052 may be stored, provided
to, or otherwise associated with a persistence store 1064, for
example, via a serialization and persistence process 1060. In
various embodiments, management micro services 1052 may be stored,
provided to, or otherwise associated with a compute element 1066,
for example, via a CI/CD (deployment and release management)
process 1062. In some embodiments, infrastructure 1006 may include
or use a container system and/or application deployment management
application, such as Kubernetes (developed by the Cloud Native
Computing Foundation).
[0119] FIG. 11 illustrates an example of an operating environment
1100 that may be representative of various embodiments. The
operating environment 1100 depicted in FIG. 11 may implement
computational model processes according to various embodiments. As
shown in FIG. 11, operating environment 1100 may include a model
management framework 1102 that includes various elements.
Non-limiting examples of elements may include data management and
persistence 1104, model evaluation and scoring management 106,
model registry and inventory 1108, model management and persistence
1110, workflow management and status trace 1112, environments
management and mapping 1114, model retraining 1116, model
deployment and release (CI/CD) 1118. In various embodiments, model
management framework 1102 may include a data streaming platform,
such as Apache Kafka.RTM. 1120 that may include stream 1122, topics
1124, and/or connector sink 1126 elements.
[0120] Model management framework 1102 may interface with data
sources 1140, data process instances 1130, modeling instances 1132,
model validation instances 1134, deployment instances 1136, and/or
data storage 1170, and/or interfaces 1150, such as management micro
services 1154. In some embodiments, deployment instance 1136 may
interact with prediction instances 1138 and/or interfaces 1150,
such as cognitive micro services.
[0121] In various embodiments, a computational model process of
operating environment 1100 may include uploading or otherwise
providing original data to model management framework 1002 (for
instance, through upload micro service and/or data streaming
service (for example, Kafka.RTM. data feed topic)), saving data
meta/version information to data streaming platform 1120 (for
instance, Kafka.RTM.) for data discovery, and save real data to
data storage (for example, a large or cloud storage solution). Data
process instances 1130 may discover new version of data from data
streaming platform 1120, process and engineering features, then
upload back to data streaming platform 1120 with new version of
meta inserted to data streaming platform 1120. Modeling instances
1132 may discover new version of models from data streaming
platform 1120, then perform testing and upload status/results.
Deployment instance 1136 may discover models and deploy prediction
instances 1138. In various embodiments, prediction instances 1138
may provide cognitive capabilities (for instance, AI, ML, and/or
the like). Model retraining may be triggered after new data has
arrived (for example, in a continuous loop).
[0122] FIG. 12 illustrates an example of an operating environment
1200 that may be representative of various embodiments. The
operating environment 1200 depicted in FIG. 12 may implement
computational model processes according to various embodiments. As
shown in FIG. 12, operating environment 1200 may include a model
management framework 1202 associated with a data streaming service
1204. In some embodiments, for example, data streaming service 1204
may be implemented with Kafka.RTM., for instance, with access
control on a topic level. Access control may include, for example,
assigning "user 1" a working set, but using "user 2" for
access.
[0123] Data streaming service 1204 may include or access a
plurality of business cases 1206a-n (use cases, problems,
objectives, business controls, and/or the like). In some
embodiments, business cases 1206a-n may include various data
elements 1250, 1252 including, without limitation, topics (for
example, the same or substantially the same as Kafka.RTM. topics),
messages, working sets, files, documents, and/or other data
elements (see, for example, FIG. 13). Non-limiting examples of data
elements may include data version topics 1250 and/or model version
topics 1252. In some embodiments, data version topics 1250 may
operate as data elements configured to store versions of data, data
meta, and/or data descriptors. In various embodiments, model
version topics 1252 may be configured to store versions of models,
model meta, event logs associated with models (for instance, model
create, update, validation, deployment, and/or the like). In some
embodiments, model management framework 1202 may be operative to
perform various functions, such as AI and/or ML functions. A
non-limiting example of a function may include classification of
documents of business cases 1206a-n.
[0124] In some embodiments, model management framework 1202 may be
associated with various services 1208. Data/model discovery 1210
may operate to determine, generate, access, and/or otherwise
interact with data process instances 1220, modeling instances 1222,
model validation instances, and/or admin user interfaces 1226. In
various embodiments, services 1208 may interact or otherwise be
associated with deployment instances 1240. In various embodiments,
deployment instances may interact with or otherwise be associated
with cognitive services code 1242 and/or tiller server 1250 and/or
a container system and/or application deployment management
application 1252. In some embodiments, model management framework
may store and/or access models to/from data storage 1230.
[0125] FIG. 13 illustrates an example of an operating environment
1300 that may be representative of various embodiments. The
operating environment 1300 depicted in FIG. 13 may implement
computational model processes, such as a data and/or model
managed/controlled via a "working set" process, according to
various embodiments. As shown in FIG. 13, operating environment
1300 may include an access control 1302 component having a working
set 1304. In various embodiments, working set 1304 may be or may be
associated with a business case, such as one of business cases
1206a-n. In various embodiments, a working set 1304 may be
categorized by use case. In various embodiments, deeper granularity
into a working set or use case may be provided, such as ECMS data
set 1, ECMS data set 2, and/or the like (for example, if different
access controls are required/desired within one use case). In some
embodiments, data sets may include public data sets associated with
public data, models, and/or the like (for instance, work2vec model
trained over wiki data, can be sharable to any user and accordingly
lower levels of access control may be applied). In other
embodiments, data sets may include case-specific (or private) data
sets (for example, business controls data sets accessible only to
business control users).
[0126] In exemplary embodiments, working set 1304 may be
implemented via a data streaming platform, such as Kafka.RTM.. In
various embodiments, working set 1304 may include a data version
topic 1306 (for example, for a current set) and/or a model version
topic 1308 (for example, for a current set). In various
embodiments, working set 1304 may be associated with various
services and may store/access data to/from storage 1340, such as
data files 1342 (for example, for a current set) and/or model files
1344 (for example, for a current set). In various embodiments, data
version topic 1330 may include or provide various data information
1320 (for example, identity of data, data type, and/or the like).
In exemplary embodiments, model version topic 1332 may include or
provide model information 1322 (for example, identity of model,
model type, and/or the like). In a Kafka.RTM. implementation, a new
working set may be created via an access control administration
service, for example, via calling "admin service" to create topics
API ("data version topic," "model version topic," and/or the like).
ACL properties may be set via an ACL setting (for instance, a
Kafka.RTM. access control setting). As new data is generated,
obtained, or otherwise accessed, the new data may be added to data
version topic 1306 amd,pde; version topic 1308 may be updated.
[0127] Computational model processes may implemewnt data and/or
model persistence according to some embodiments. For example, data
and/or model files may include relatively larger (or larger
compared with data meta) objects with low frequency updates, which
may be handled via a big data solution. Data and/or model meta may
include data defining attributes, such as small data objects with
relatively high frequency of updates (for instance, stored in a
data version topic 1320 or model version topic 1322,
respectively).
[0128] In some embodiments, data persistence for data may include
uploading data with meta data framework (for instance, upload micro
service and/or data feeding topic). The framework may perform an
access control check and save data files to storage and get a
storage key/link. The framework may generate complete meta and set
it to data version topic, which may be discovered by listing the
data version topic. In various embodiments, persistence for models
may include a modeling instance generating a model and uploading
the model with meta information to framework (for instance, upload
micro service and/or model feeding topic). The framework may
perform an access control check and save data model to storage and
get a storage key/link. The framework may generate complete meta
and set it to model version topic, which may be discovered by
listing the model version topic.
[0129] In some embodiments, data/model discovery may therefore
include data/model version topics with all associated meta
information and/or access control. Listing the data/model version
topic may provide for discovery of all data/model for a current
working set. A message selector/filter may be applied to access
data/model of interest.
[0130] Included herein are one or more logic flows representative
of exemplary methodologies for performing novel aspects of the
disclosed architecture. While, for purposes of simplicity of
explanation, the one or more methodologies shown herein are shown
and described as a series of acts, those skilled in the art will
understand and appreciate that the methodologies are not limited by
the order of acts. Some acts may, in accordance therewith, occur in
a different order and/or concurrently with other acts from that
shown and described herein. For example, those skilled in the art
will understand and appreciate that a methodology could
alternatively be represented as a series of interrelated states or
events, such as in a state diagram. Moreover, not all acts
illustrated in a methodology may be required for a novel
implementation.
[0131] A logic flow may be implemented in software, firmware,
hardware, or any combination thereof. In software and firmware
embodiments, a logic flow may be implemented by computer executable
instructions stored on a non-transitory computer readable medium or
machine readable medium. Embodiments are not limited in this
context.
[0132] FIG. 14 illustrates an embodiment of a logic flow 1400.
Logic flow 1400 may be representative of some or all of the
operations executed by one or more embodiments described herein,
such as computing device 701. In some embodiments, logic flow 1400
may be representative of some or all of the operations of a
computational model process.
[0133] Logic flow 1400 may create a new working set at block 1402.
For example, a new working set may be created for a business case
(for instance, via calling "admin service" or a similar process in
model management framework). A data version topic and a model
version topic may be created for the working set. In some
embodiments, data and/or model storage folders may be created for
the new working set.
[0134] At block 1404, logic flow 1400 may perform a data upload
service. For example, an original rank training may be uploaded
(for example, via calling a "data upload service") in the model
management framework. A data version may be generated and the data
meta may be prepared. The data file may be stored in a folder and a
new version of data meta may be inserted in the data version topic.
At block 1406, logic flow 1400 may access data versions. For
example, all data versions and meta from data version topic may be
accessed (for instance, via retrieving all messages from data
version topic through a Kafka.RTM. API in a Kafka.RTM.
implementation). Logic flow 1400 may model upload service at 1408.
For example, a pretrained model, such as a paragraph2vector (P2V)
model may be uploaded (for instance, via calling a "model upload
service" in the model management framework). The model version may
be created and the model meta prepared. A new version of model meta
may be inserted in the model version topic.
[0135] Logic flow 1400 may process training data at block 1410. For
example, the rank training data may be processed (for instance, via
converting original data to vector numbers by P2V model). The
original rank training file (for instance, a comma delimited or
Microsoft.RTM. Excel.RTM. file) may be downloaded from the
framework and the P2V model may be downloaded from the framework.
In some embodiments, data may be read from the original rank
training file, and records may be converted to vectors to generate
a vectorized file. The new vectorized file may be uploaded with
meta and inserted as a new version of data meta to the data version
topic.
[0136] At block 1412, logic flow 1400 may train a ranked model. For
example, the vectorized training file may be downloaded from the
framework. A training process may be performed to generate a new
rank model. The new rank model file may be uploaded with meta and
inserted as a new model meta version into the model version topic.
Logic flow 1400 may access model versions at block 1414. For
example, all model versions may be viewed and meta from model
version topic (for instance, via a data repository platform, such
as Kafka.RTM.), such as by retrieving all messages from the model
version topic.
[0137] Logic flow 1400 may validate the model at block 1416. For
example, all model versions in the model version topic may be
listed and one or more selected models may be downloaded from the
framework. The model may be validated according to various
embodiments. The model meta may be updated and a new version of the
model meta may be inserted into the model version topic. At block
1418, logic flow 1400 may deploy the model. For example, all model
versions in the model version topic may be listed. The selected
model for deployment may be downloaded and the model deployed. The
model meta may be updated and a new version of the model meta may
be inserted into the model version topic.
[0138] FIG. 15 illustrates an embodiment of a logic flow 1500.
Logic flow 1500 may be representative of some or all of the
operations executed by one or more embodiments described herein,
such as computing device 701. In some embodiments, logic flow 1500
may be representative of some or all of the operations of creating
a working set within a computational model process.
[0139] Logic flow 1500 may call an admin service at block 1502 and
check access control for the admin service at block 1504. At block
1506, logic flow 1500 may created storage folders for the new
working set. Logic flow 1500 may create data version topic and
assign access controls at block 1508 and create model version topic
and assign access controls at block 1510 to create the working set
at block 1512.
[0140] FIG. 16 illustrates an embodiment of a logic flow 1600.
Logic flow 1600 may be representative of some or all of the
operations executed by one or more embodiments described herein,
such as computing device 701. In some embodiments, logic flow 1600
may be representative of some or all of the operations of uploading
original (rank) training data within a computational model
process.
[0141] Logic flow 1600 may call data upload service at block 1602,
check access control for current upload at block 1604, save the
data file in storage at block 1606, generate data version and send
a version of data meta to topic at block 1608, call upload service
at block 1610, and list the data version and meta in data version
topic at block 1612.
[0142] FIG. 17 illustrates an embodiment of a logic flow 1700.
Logic flow 1700 may be representative of some or all of the
operations executed by one or more embodiments described herein,
such as computing device 701. In some embodiments, logic flow 1700
may be representative of some or all of the operations of uploading
a pre-trained (P2V) model within a computational model process.
[0143] Logic flow 1700 may call model upload service at block 1702,
check access control for current upload at block 1704, save the
model file in storage at block 1706, generate model version and
send a version of model meta to topic at block 1708, call upload
service at block 1710, and list the model version and meta in model
version topic at block 1712.
[0144] FIG. 18 illustrates an embodiment of a logic flow 1800.
Logic flow 1800 may be representative of some or all of the
operations executed by one or more embodiments described herein,
such as computing device 701. In some embodiments, logic flow 1800
may be representative of some or all of the operations of
processing training data (for example, vectorization) within a
computational model process.
[0145] Logic flow 1800 may download the training file and (P2V)
model file at block 1202, initialize the (P2V) model at block 1804,
get training records at block 1806, perform vectorization by using
(P2V) model and generate vectorized file at block 1808, upload
vectorized file with meta at block 1810, and list the data version
and meta in data version topic at block 1812.
[0146] FIG. 19 illustrates an embodiment of a logic flow 1900.
Logic flow 1900 may be representative of some or all of the
operations executed by one or more embodiments described herein,
such as computing device 701. In some embodiments, logic flow 1900
may be representative of some or all of the operations of training
a rank model within a computational model process.
[0147] Logic flow 1900 may download vectorized rank training data
at block 1902, run a training process at block 1904, generate new
rank model file at block 1906, upload new rank model file with meat
at block 1908, and list the model version and meta in the model
version topic at block 1910.
[0148] FIG. 20 illustrates an embodiment of a logic flow 2000.
Logic flow 2000 may be representative of some or all of the
operations executed by one or more embodiments described herein,
such as computing device 701. In some embodiments, logic flow 2000
may be representative of some or all of the operations of
validating a rank model within a computational model process.
[0149] Logic flow 2000 may list the model version from the model
version topic at 2002, select the model for validation and download
the model at block 2004, validate the model at block 2006, update
the model meta for validation status/results at block 2008, and
list the model version and meta in the model version topic at block
2010.
[0150] FIG. 21 illustrates an embodiment of a logic flow 2100.
Logic flow 2100 may be representative of some or all of the
operations executed by one or more embodiments described herein,
such as computing device 701. In some embodiments, logic flow 2100
may be representative of some or all of the operations of deploying
a rank model within a computational model process.
[0151] Logic flow 2100 may list the model version from the model
version topic at block 2102, select the model for deployment and
download the model at block 2104, deploy the model at block 2106,
update the model meta for deployment status/meta at block 2108, and
list the model version and meta in the model version topic at block
2110.
[0152] FIG. 22 illustrates an embodiment of a logic flow 2200.
Logic flow 2200 may be representative of some or all of the
operations executed by one or more embodiments described herein,
such as computing device 701. In some embodiments, logic flow 2200
may be representative of some or all of the operations of auto rank
training with new data and model decay prevention within a
computational model process.
[0153] Logic flow 2200 may start a data version listener on data
version topic at block 2202, state a data process logic to generate
a new version of vectorized rank data at block 2204, provide new
data meta to data version topic at block 2206, obtain new data
notification and download the new data via a listener at block
2208, trigger a new round of rank training via the listener and
generate a rank model at block 2210, update the new rank model at
block 2212, and list the model version and meta in the model
version topic at block 2214.
[0154] FIG. 23 illustrates an embodiment of a logic flow 2300.
Logic flow 2300 may be representative of some or all of the
operations executed by one or more embodiments described herein,
such as computing device 701. In some embodiments, logic flow 2300
may be representative of some or all of the operations of automated
(P2V) model training with new data within a computational model
process.
[0155] Logic flow 2300 may start a data version listener on data
version topic at block 2302, upload a new (P2V) training data set
at block 2304, put new data meta into data version topic at block
2306, obtain new data notification and download the new data via a
listener at block 2308, trigger a new round of (P2V) training and
generate a (P2V) model via the listener at block 2310, update the
new (P2V) model at block 2312, and list the model version and meta
in the model version topic at block 2314.
[0156] Some embodiments may be described using the expression "one
embodiment" or "an embodiment" along with their derivatives. These
terms mean that a particular feature, structure, or characteristic
described in connection with the embodiment is included in at least
one embodiment. The appearances of the phrase "in one embodiment"
in various places in the specification are not necessarily all
referring to the same embodiment. Further, some embodiments may be
described using the expression "coupled" and "connected" along with
their derivatives. These terms are not necessarily intended as
synonyms for each other. For example, some embodiments may be
described using the terms "connected" and/or "coupled" to indicate
that two or more elements are in direct physical or electrical
contact with each other. The term "coupled," however, may also mean
that two or more elements are not in direct contact with each
other, but yet still co-operate or interact with each other.
[0157] With general reference to notations and nomenclature used
herein, the detailed descriptions herein may be presented in terms
of program procedures executed on a computer or network of
computers. These procedural descriptions and representations are
used by those skilled in the art to most effectively convey the
substance of their work to others skilled in the art.
[0158] A procedure is here, and generally, conceived to be a
self-consistent sequence of operations leading to a desired result.
These operations are those requiring physical manipulations of
physical quantities. Usually, though not necessarily, these
quantities take the form of electrical, magnetic or optical signals
capable of being stored, transferred, combined, compared, and
otherwise manipulated. It proves convenient at times, principally
for reasons of common usage, to refer to these signals as bits,
values, elements, symbols, characters, terms, numbers, or the like.
It should be noted, however, that all of these and similar terms
are to be associated with the appropriate physical quantities and
are merely convenient labels applied to those quantities.
[0159] Further, the manipulations performed are often referred to
in terms, such as adding or comparing, which are commonly
associated with mental operations performed by a human operator. No
such capability of a human operator is necessary, or desirable in
most cases, in any of the operations described herein which form
part of one or more embodiments. Rather, the operations are machine
operations. Useful machines for performing operations of various
embodiments include general purpose digital computers or similar
devices.
[0160] Various embodiments also relate to apparatus or systems for
performing these operations. This apparatus may be specially
constructed for the required purpose or it may comprise a
general-purpose computer as selectively activated or reconfigured
by a computer program stored in the computer. The procedures
presented herein are not inherently related to a particular
computer or other apparatus. Various general-purpose machines may
be used with programs written in accordance with the teachings
herein, or it may prove convenient to construct more specialized
apparatus to perform the required method steps. The required
structure for a variety of these machines will appear from the
description given.
[0161] In addition, in the foregoing Detailed Description, it can
be seen that various features are grouped together in a single
embodiment for the purpose of streamlining the disclosure. This
method of disclosure is not to be interpreted as reflecting an
intention that the claimed embodiments require more features than
are expressly recited in each claim. Rather, as the following
claims reflect, inventive subject matter lies in less than all
features of a single disclosed embodiment. Thus the following
claims are hereby incorporated into the Detailed Description, with
each claim standing on its own as a separate embodiment. In the
appended claims, the terms "including" and "in which" are used as
the plain-English equivalents of the respective terms "comprising"
and "wherein," respectively. Moreover, the terms "first," "second,"
"third," and so forth, are used merely as labels, and are not
intended to impose numerical requirements on their objects.
[0162] What has been described above includes examples of the
disclosed architecture. It is, of course, not possible to describe
every conceivable combination of components and/or methodologies,
but one of ordinary skill in the art may recognize that many
further combinations and permutations are possible.
* * * * *