Patent Application
20220327331
Software services based on Artificial Intelligence (AI) technology often relies on training data. This implies that when the service is deployed on a Customer's data, if there is a mismatch between the Customer's data and the training data then the service may not perform well.
Embodiments relate to defining data for evaluating artificial intelligence (AI) services. One embodiment provides for a method for evaluation of an AI service, the method includes partitioning, by a processor, data into in-domain data and out-of-domain data. The processor defines held-out data from both of the in-domain data and the out-of-domain data for evaluation by each of domain and sub-domain based on building a taxonomy of both domains and sub-domains for the AI service. The held-out data is excluded from training data used for training the AI service. The processor further determines distribution underlying performance metrics for the held-out data using bootstrap validation processing. The processor also determines performance guarantees for multiple settings conditioned on multiple characteristics of an application scenario for the held-out data of the taxonomy based on the underlying performance metrics.
These and other features, aspects and advantages of the present invention will become understood with reference to the following description, appended claims and accompanying figures.
The descriptions of the various embodiments have been presented for purposes of illustration, but are not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein was chosen to best explain the principles of the embodiments, the practical application or technical improvement over technologies found in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
It is understood in advance that although this disclosure includes a detailed description of cloud computing, implementation of the teachings recited herein are not limited to a cloud computing environment. Rather, embodiments of the present invention are capable of being implemented in conjunction with any other type of computing environment now known or later developed.
Embodiments relate to evaluating quality of AI services and determining performance guarantees. One embodiment provides a method for evaluation of an AI service, the method includes partitioning, by a processor, data into in-domain data and out-of-domain data. The processor defines held-out data from the in-domain data and the out-of-domain data for evaluation by domain and sub-domain based on building a taxonomy of domains and sub-domains for the AI service. The processor further determines distribution underlying performance metrics for the held-out data using statistical processing. The processor also determines performance guarantees for multiple settings conditioned on multiple characteristics of an application scenario for the held-out data of the taxonomy based on the underlying performance metrics. The processor further provides confidence intervals based on the performance guarantees.
Many modern software applications rely on AI, for example automated driving, extraction from PDFs, etc. AI algorithms often rely on models built from training data. A mismatch between training data and unseen data, on which the AI service is later deployed, potentially leads to poor results. In one embodiment, an evaluation process is performed that is more objective than conventional systems, quantifies performance in various settings such as performance on different data: within domain, out-of-domain, and differing customer requirements: heavy users and light users; and provides more informative guarantees on performance of each setting. In one embodiment, the performance guarantees may also be used to draw up contracts between the provider of the service and a user.
Conventional evaluation services evaluate AI services on validation data. Compared to the complete data set U, validation data is typically small. The conventional systems have the following drawbacks: the metrics produced may not have any bearing on unseen data; on different sets of unseen data, the same AI service may produce results of differing quality; since developers may be able to “see” validation data, this raises the possibility of “contamination;” and computed metrics may be biased since validation data is not truly held out.
Cloud computing is a model of service delivery for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, network bandwidth, servers, processing, memory, storage, applications, virtual machines (VMs), and services) that can be rapidly provisioned and released with minimal management effort or interaction with a provider of the service. This cloud model may include at least five characteristics, at least three service models, and at least four deployment models.
Characteristics are as follows:
On-demand self-service: a cloud consumer can unilaterally provision computing capabilities, such as server time and network storage, as needed and automatically, without requiring human interaction with the service's provider.
Broad network access: capabilities are available over a network and accessed through standard mechanisms that promote use by heterogeneous, thin or thick client platforms (e.g., mobile phones, laptops, and PDAs).
Resource pooling: the provider's computing resources are pooled to serve multiple consumers using a multi-tenant model, with different physical and virtual resources dynamically assigned and reassigned according to demand. There is a sense of location independence in that the consumer generally has no control or knowledge over the exact location of the provided resources but may be able to specify location at a higher level of abstraction (e.g., country, state, or data center).
Rapid elasticity: capabilities can be rapidly and elastically provisioned and, in some cases, automatically, to quickly scale out and rapidly released to quickly scale in. To the consumer, the capabilities available for provisioning often appear to be unlimited and can be purchased in any quantity at any time.
Measured service: cloud systems automatically control and optimize resource use by leveraging a metering capability at some level of abstraction appropriate to the type of service (e.g., storage, processing, bandwidth, and active consumer accounts). Resource usage can be monitored, controlled, and reported, thereby providing transparency for both the provider and consumer of the utilized service.
Service Models are as follows:
Software as a Service (SaaS): the capability provided to the consumer is the ability to use the provider's applications running on a cloud infrastructure. The applications are accessible from various client devices through a thin client interface, such as a web browser (e.g., web-based email). The consumer does not manage or control the underlying cloud infrastructure including network, servers, operating systems, storage, or even individual application capabilities, with the possible exception of limited consumer-specific application configuration settings.
Platform as a Service (PaaS): the capability provided to the consumer is the ability to deploy onto the cloud infrastructure consumer-created or acquired applications created using programming languages and tools supported by the provider. The consumer does not manage or control the underlying cloud infrastructure including networks, servers, operating systems, or storage, but has control over the deployed applications and possibly application-hosting environment configurations.
Infrastructure as a Service (IaaS): the capability provided to the consumer is the ability to provision processing, storage, networks, and other fundamental computing resources where the consumer is able to deploy and run arbitrary software, which can include operating systems and applications. The consumer does not manage or control the underlying cloud infrastructure but has control over operating systems, storage, deployed applications, and possibly limited control of select networking components (e.g., host firewalls).
Deployment Models are as follows:
Private cloud: the cloud infrastructure is operated solely for an organization. It may be managed by the organization or a third party and may exist on-premises or off-premises.
Community cloud: the cloud infrastructure is shared by several organizations and supports a specific community that has shared concerns (e.g., mission, security requirements, policy, and compliance considerations). It may be managed by the organizations or a third party and may exist on-premises or off-premises.
Public cloud: the cloud infrastructure is made available to the general public or a large industry group and is owned by an organization selling cloud services.
Hybrid cloud: the cloud infrastructure is a composition of two or more clouds (private, community, or public) that remain unique entities but are bound together by standardized or proprietary technology that enables data and application portability (e.g., cloud bursting for load balancing between clouds).
A cloud computing environment is a service oriented with a focus on statelessness, low coupling, modularity, and semantic interoperability. At the heart of cloud computing is an infrastructure comprising a network of interconnected nodes.
Referring now to
Referring now to
Hardware and software layer 60 includes hardware and software components. Examples of hardware components include: mainframes 61; RISC (Reduced Instruction Set Computer) architecture based servers 62; servers 63; blade servers 64; storage devices 65; and networks and networking components 66. In some embodiments, software components include network application server software 67 and database software 68.
Virtualization layer 70 provides an abstraction layer from which the following examples of virtual entities may be provided: virtual servers 71; virtual storage 72; virtual networks 73, including virtual private networks; virtual applications and operating systems 74; and virtual clients 75.
In one example, a management layer 80 may provide the functions described below. Resource provisioning 81 provides dynamic procurement of computing resources and other resources that are utilized to perform tasks within the cloud computing environment. Metering and pricing 82 provide cost tracking as resources are utilized within the cloud computing environment and billing or invoicing for consumption of these resources. In one example, these resources may comprise application software licenses. Security provides identity verification for cloud consumers and tasks as well as protection for data and other resources. User portal 83 provides access to the cloud computing environment for consumers and system administrators. Service level management 84 provides cloud computing resource allocation and management such that required service levels are met. Service Level Agreement (SLA) planning and fulfillment 85 provide pre-arrangement for, and procurement of, cloud computing resources for which a future requirement is anticipated in accordance with an SLA.
Workloads layer 90 provides examples of functionality for which the cloud computing environment may be utilized. Examples of workloads and functions which may be provided from this layer include: mapping and navigation 91; software development and lifecycle management 92; virtual classroom education delivery 93; data analytics processing 94; transaction processing 95; and evaluating quality of AI services and determining performance guarantees processing 96. As mentioned above, all of the foregoing examples described with respect to
It is understood all functions of one or more embodiments as described herein may be typically performed by the processing system 300 (
It is reiterated that although this disclosure includes a detailed description on cloud computing, implementation of the teachings recited herein are not limited to a cloud computing environment. Rather, the embodiments of the present invention may be implemented with any type of clustered computing environment now known or later developed.
In use, the gateway 301 serves as an entrance point from the remote networks 302 to the proximate network 308. As such, the gateway 301 may function as a router, which is capable of directing a given packet of data that arrives at the gateway 301, and a switch, which furnishes the actual path in and out of the gateway 301 for a given packet.
Further included is at least one data server 314 coupled to the proximate network 308, which is accessible from the remote networks 302 via the gateway 301. It should be noted that the data server(s) 314 may include any type of computing device/groupware. Coupled to each data server 314 is a plurality of user devices 316. Such user devices 316 may include a desktop computer, laptop computer, handheld computer, printer, and/or any other type of logic-containing device. It should be noted that a user device 316 may also be directly coupled to any of the networks in some embodiments.
A peripheral 320 or series of peripherals 320, e.g., facsimile machines, printers, scanners, hard disk drives, networked and/or local storage units or systems, etc., may be coupled to one or more of the networks 304, 306, 308. It should be noted that databases and/or additional components may be utilized with, or integrated into, any type of network element coupled to the networks 304, 306, 308. In the context of the present description, a network element may refer to any component of a network.
According to some approaches, methods and systems described herein may be implemented with and/or on virtual systems and/or systems, which emulate one or more other systems, such as a UNIX system that emulates an IBM z/OS environment, a UNIX system that virtually hosts a MICROSOFT WINDOWS environment, a MICROSOFT WINDOWS system that emulates an IBM z/OS environment, etc. This virtualization and/or emulation may be implemented through the use of VMWARE software in some embodiments.
In one example, the workstation may have resident thereon an operating system, such as the MICROSOFT WINDOWS Operating System (OS), a MAC OS, a UNIX OS, etc. In one embodiment, the system 400 employs a POSIX® based file system. It will be appreciated that other examples may also be implemented on platforms and operating systems other than those mentioned. Such other examples may include operating systems written using JAVA, XML, C, and/or C++ language, or other programming languages, along with an object oriented programming methodology. Object oriented programming (OOP), which has become increasingly used to develop complex applications, may also be used.
In one embodiment, system 500, defines and organizes held-out data for evaluation by domain/sub-domain by building and maintaining a taxonomy. The workflow achieves objective evaluation, provides performance guarantees conditioned on various characteristics of the application scenario such as data characteristics (in-domain versus out-of-domain), etc., also supports providing performance guarantees conditioned on various user-characteristics (such as number of application programming interface (API) calls), and provides informative performance guarantees for each setting.
In one embodiment, block 610 defines domains and sub-domains as an ongoing enhancement. The defined domains and sub-domains are stored in the data store 605 (e.g., memory device(s), disk drive(s), storage system, etc.). Block 620 provides labeled data N for sub-domain S for input to block 630. This is an ongoing process with a manual evaluation and curation. Existing labeled data for S is input to block 630, where data selection for sub-domain S occurs. Block 630 determined hold out labeled data (e.g., datasets model builders cannot access) for evaluation (stored in the data store 605). In one embodiment, standard techniques to perform block 630 include stratified sampling. In settings where classification models are required, stratified sampling allows one to subdivide data so that the class proportions remain intact. For instance, if N contains 30 positive labels and 70 negative labels, then stratified sampling may be used to divide it into a training set and held out set so that both sets contain 30:70 positive:negative labels. In certain cases, even stratified sampling may not be needed. For instance, consider N that contains documents which contain sentences that need to be labeled. If it is assumed that each document contains sentences labeled with positive and negative labels in roughly the same proportion (e.g., each document contains 30% sentences labeled positive, and 70% sentences labeled negative) then simply assigning documents randomly to the training and held out set may suffice.
From block 630, labeled data for model training are communicated to block 640 for model training (e.g., AI, models, deep learning models, machine learning, etc.). While the kind of models trained and the kind of model training algorithms are not critical for block 640, some example embodiments include deep learning models (convolutional neural networks and recurrent neural networks) trained with mini-batch stochastic gradient methods and convex machine learning models that are more amenable to full-batch gradient and second-order optimization techniques which access the hessian of the objective function to speed up convergence.
Using data from the data store 605, block 650 provides data selection for out of domain S. The out of domain data for S is input to block 660 that performs evaluation services. In one embodiment, examples of evaluation services may include cross validation, which divides the (held out) data into K equal parts and estimates the goodness of predictions of each part (e.g., in terms of precision and recall, but not limited to these metrics). Given these results, one embodiment may gain insight into the distribution of the metrics by measuring average and standard deviation. Another evaluation method that may be used in block 660 is a bootstrap that actually provides statistical guarantees beyond the held out data. As opposed to cross validation, bootstrap validation repeatedly creates “new” held out data by randomly sampling from the provided held out data. In one embodiment, block 660 may make copies of data items or does not select a data item for inclusion into the bootstrap. By measuring goodness of predictions from each bootstrap, one embodiment may gain insight into the distribution underlying each evaluation metric. By dropping the lowest and highest 2.5% values of a particular metric, one embodiment may determine the 95% confidence interval.
Raw data from the evaluation services block 660 are communicated to block 670 that includes services (e.g., PDF, natural language processing (NLP), etc.). The service block 670 provides service output data to block 660 for evaluation services. The output from block 660 includes a quality evaluation for in-domain data and out-of-domain data. A representation, such as JSON 680 (or any other format, such as comma-separated values (CSV), etc.) may be determined from the output of block 660.
In one embodiment, in process 1100 the performance metrics comprises precision information and recall information. In other embodiments, other evaluation metrics may be used (e.g., F-measure, accuracy, etc.). In one embodiment, the held-out data is: protected by enterprise-level security, never accessed by a developer of the AI service, and ensures uncontaminated development of the AI service.
In one embodiment, process 1100 may include partitioning, by the processor, new data into representative portions. One representative portion is provided to developers and a second representative portion is defined as new held-out data. Process 1100 may provide that the multiple characteristics comprise data characteristics and user characteristics.
In one embodiment, process 1100 may include that upon evaluating the AI service, a first portion of nodes in the taxonomy provides the in-domain data, and second portion of nodes in the taxonomy provide the out-of-domain data. In process 1100, a portion of the in-domain data and the out-of-domain data undefined as held-out data is communicated as training data for model training for the AI service.
As will be appreciated by one skilled in the art, aspects of the present invention may be embodied as a system, method or computer program product. Accordingly, aspects of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.) or an embodiment combining software and hardware aspects that may all generally be referred to herein as a “circuit,” “module” or “system.” Furthermore, aspects of the present invention may take the form of a computer program product embodied in one or more computer readable medium(s) having computer readable program code embodied thereon.
Any combination of one or more computer readable medium(s) may be utilized. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++or the like and conventional procedural programming languages, such as the “C” programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider).
Aspects of the present invention are described below with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer readable medium that can direct a computer, other programmable data processing apparatus, or other devices to function in a particular manner, such that the instructions stored in the computer readable medium produce an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.
The computer program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks.
The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts or carry out combinations of special purpose hardware and computer instructions.
References in the claims to an element in the singular is not intended to mean “one and only” unless explicitly so stated, but rather “one or more.” All structural and functional equivalents to the elements of the above-described exemplary embodiment that are currently known or later come to be known to those of ordinary skill in the art are intended to be encompassed by the present claims. No claim element herein is to be construed under the provisions of 35 U.S.C. section 112, sixth paragraph, unless the element is expressly recited using the phrase “means for” or “step for.”
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises” and/or “comprising,” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
The corresponding structures, materials, acts, and equivalents of all means or step plus function elements in the claims below are intended to include any structure, material, or act for performing the function in combination with other claimed elements as specifically claimed. The description of the present invention has been presented for purposes of illustration and description, but is not intended to be exhaustive or limited to the invention in the form disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the invention. The embodiment was chosen and described in order to best explain the principles of the invention and the practical application, and to enable others of ordinary skill in the art to understand the invention for various embodiments with various modifications as are suited to the particular use contemplated.
| Number | Date | Country | |
|---|---|---|---|
| Parent | 16123822 | Sep 2018 | US |
| Child | 17809537 | US |
