Patent Application
20230289620
Machine learning (ML) provides computers with an ability to learn, or continue learning, without being pre-programmed. Machine learning utilizes algorithms that learn from data and create insights based on the data, such as making predictions or decisions.
Training data in machine learning is the data used to train a model to make a prediction and solve a problem, provide relevant recommendations, perform an action, etc. Supervised learning refers to the task of inducing a learning function from a set of labeled data examples so the function can map between the input (input variable(s)) and the output (output variable(s)) in the training examples. After training, the created model should be able to generalize and correctly predict output(s) for unseen datapoints.
Certain shortcomings of the prior art are overcome and additional advantages are provided through the provision, in one or more aspects, of a computer program product for facilitating processing within a computing environment. The computer program product includes one or more computer-readable storage media and program instructions embodied therewith. The program instructions are readable by a processing circuit to cause the processing circuit to perform a method. The method includes obtaining training data for use in training a machine learning model, where the training data includes factual data relating to a particular application, and obtaining with reference to the training data, unlabeled counterfactual data for the particular application. Further, the method includes imputing one or more labels to the unlabeled counterfactual data, using domain knowledge for the particular application, to obtain imputed counterfactual data. The domain knowledge includes one or more application domain properties. Further, the method includes training the machine learning model using the training data and the imputed counterfactual data to facilitate generating machine learning model predictions for the particular application in accordance with the one or more application domain properties.
Computer systems and computer-implemented methods relating to one or more aspects are also described and claimed herein. Further, services relating to one or more aspects are also described and may be claimed herein.
Additional features and advantages are realized through the techniques described herein. Other embodiments and aspects are described in detail herein and are considered a part of the claimed aspects.
One or more aspects are particularly pointed out and distinctly claimed as examples in the claims at the conclusion of the specification. The foregoing and objects, features, and advantages of one or more aspects are apparent from the following detailed description taken in conjunction with the accompanying drawings in which:
The accompanying figures, which are incorporated in and form a part of this specification, further illustrate the present invention and, together with this detailed description of the invention, serve to explain aspects of the present invention. Note in this regard that descriptions of well-known systems, devices, processing techniques, etc., are omitted so as to not unnecessarily obscure the invention in detail. It should be understood, however, that the detailed description and this specific example(s), while indicating aspects of the invention, are given by way of illustration only, and not limitation. Various substitutions, modifications, additions, and/or other arrangements, within the spirit or scope of the underlying inventive concepts will be apparent to those skilled in the art from this disclosure. Note further that numerous inventive aspects or features are disclosed herein, and unless inconsistent, each disclosed aspect or feature is combinable with any other disclosed aspect or feature as desired for a particular application of the concepts disclosed herein.
Note also that illustrative embodiments are described below using specific code, designs, architectures, protocols, layouts, schematics, or tools only as examples, and not by way of limitation. Furthermore, the illustrative embodiments are described in certain instances using particular software, tools, or data processing environments only as example for clarity of description. The illustrative embodiments can be used in conjunction with other comparable or similarly purposed structures, systems, applications, or architectures. One or more aspects of an illustrative embodiment can be implemented in hardware, software, or a combination thereof.
As understood by one skilled in the art, program code, as referred to in this application, can include both software and hardware. For example, program code in certain embodiments of the present invention can utilize a software-based implementation of the functions described, while other embodiments can include fixed function hardware. Certain embodiments combine both types of program code. Examples of program code, also referred to as one or more programs, are depicted in
As noted, machine learning provides computers with the ability to continually learn without being preprogrammed. Machine learning utilizes algorithms that learn from data and create insights based on the data, such as provide relevant recommendations, solve problems, perform actions, etc. A machine learning algorithm needs data to learn from. Supervised learning refers to the task of inducing a learning function from a set of labeled examples so the function can map between the input (input variable(s)) and the output (output variable(s)) in the training examples. After training, the created model should be able to generalize and correctly predict an output for unseen datapoints.
One of the challenges when applying machine learning models to real applications is the need to ensure that the learned function or model is in agreement with one or more known application domain properties, such as one or more monotonic-type properties or relationships with respect to certain inputs. Capturing such relationships during training of the machine learning model is advantageous, and often necessary for operational feasibility of the model.
By way of example, a monotonic-type property might relate to one or more attributes associated with a particular application. For instance, where a value of an attribute increases, such as computing system performance, then everything else being equal in a multi-system environment, demand should increase for the higher performing system. Similarly, where system performance decreases, then demand should decrease for the lower performing system.
While being powerful and flexible at uncovering complex latent relationships, deep neural networks (DNNs), which are often complex, heavily engineered, opaque systems, can fail to generalize a known application domain property or relationship correctly using available labeled training data, leading to suboptimal model predictions. One approach to addressing this is to utilize simpler models cable of capturing monotonicity. Unfortunately, this can result in a loss of prediction accuracy (such as constraining the sign of a coefficient for a linear model). In another, model-specific approach, monotonicity can be configured into particular DNNs with a specific network structure. However, such model-specific approaches can be restrictive, complex to implement and also perform poorly empirically. Thus, to ensure that an application domain property, such as a monotonic relationship, is correctly captured, certain applications in practice have had to fall back to simpler and less accurate models (e.g., linear models), where such attributes or constraints can be more readily incorporated.
Advantageously, disclosed herein is a machine learning model-agnostic approach to augmenting a training data set to encourage inclusion of a desired application domain property, such as one or more monotonic-type properties or relationships, in predictions of the trained model. In one or more embodiments, unlabeled counterfactual data is leveraged by assigning or imputing where possible labels to the counterfactual data that are known to obey (for instance) the desired monotonicity, and to add these imputed counterfactual data points to the existing training data set. (Note that as used herein, imputing refers to assigning or ascribing by inference a value or label.) This augmentation approach can be used in a variety of applications, including, for instance, as data preprocessing, or be integrated into a self-training procedure for counterfactual inference, such as described herein. Prior to describing such machine learning model-agnostic approaches further, machine learning is further discussed below in the context of the embodiments of
In one or more implementations, computing resource(s) 110 houses and/or executes program code 112 configured to perform methods in accordance with one or more aspects of the present invention. By way of example, computing resource(s) 110 can be a server or other computing-system-implemented resource(s). Further, for illustrative purposes only, computing resource(s) 110 in
Briefly described, in one embodiment, computing resource(s) 110 can include one or more processors, for instance, central processing units (CPUs). Also, the processor(s) can include functional components used in the integration of program code, such as functional components to fetch program code from locations, such as cache or main memory, decode program code, and execute program code, access memory for instruction execution, and write results of the executed instructions or code. The processor(s) can also include a register(s) to be used by one or more of the functional components. In one or more embodiments, the computing resource(s) can include memory, input/output, a network interface, and storage, which can include and/or access, one or more other computing resources and/or databases, as required to implement the machine learning processing described herein. The components of the respective computing resource(s) can be coupled to each other via one or more buses and/or other connections. Bus connections can be one or more of any of several types of bus structures, including a memory bus or a memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus, using any of a variety of architectures. By way of example, but not limitation, such architectures can include the Industry Standard Architecture (ISA), the micro-channel architecture (MCA), the enhanced ISA (EISA), the Video Electronic Standard Association (VESA), local bus, and peripheral component interconnect (PCI). Examples of a computing resource(s) or a computer system(s) which can implement one or more aspects disclosed herein are described further below with reference to
As noted, program code 112 executes, in one implementation, a cognitive engine 114 which includes one or more machine learning agents 116 that facilitate training one or more machine learning models 118. The machine learning models are trained using one or more training datasets that can include a variety of types of data, depending on the model and the data sources. In one or more embodiments, program code 112 executing on one or more computing resources 110 applies machine learning algorithms of machine learning agent 116 to generate and train the model(s), which the program code then utilizes to make a prediction, perform a skill (e.g., provide a solution, make a recommendation, perform an action, etc.). In an initialization or learning stage, program code 112 trains one or more machine learning models 118 using a received or obtained training dataset that can include, in one or more embodiments, labeled factual data and unlabeled counterfactual data, such as described herein.
Some embodiments of the present invention can utilize IBM Watson® as learning agent. IBM Watson® is a registered trademark of International Business Machines Corporation, Armonk, N.Y., USA. In embodiments of the present invention, the respective program code can interface with IBM Watson® application program interfaces (APIs) to perform machine learning analysis of obtained data. In some embodiments of the present invention, the respective program code can interface with the application programming interfaces (APIs) that are part of a known machine learning agent, such as the IBM Watson® application programming interface (API), a product of International Business Machines Corporation, to determine impacts of data on the machine learning model, and to update the model, accordingly.
In some embodiments of the present invention, certain of the APIs of the IBM Watson® API include a machine learning agent (e.g., learning agent) that includes one or more programs, including, but not limited to, natural language classifiers, Retrieve-and-Rank (i.e., a service available through the IBM Watson® developer cloud that can surface the most relevant information from a collection of documents), concepts/visualization insights, tradeoff analytics, document conversion, natural language processing, and/or relationship extraction. In an embodiment of the present invention, one or more programs can be provided to analyze data obtained by the program code across various sources utilizing one or more of, for instance, a natural language classifier, Retrieve-and-Rank APIs, and tradeoff analytics APIs.
In some embodiments of the present invention, the program code utilizes a neural network to analyze training data and/or collected data to generate an operational model or machine learning model. Neural networks are a programming paradigm which enable a computer to learn from observational data. This learning is referred to as deep learning, which is a set of techniques for learning in neural networks. Neural networks, including modular neural networks, are capable of pattern (e.g., state) recognition with speed, accuracy, and efficiency, in situations where datasets are mutual and expansive, including across a distributed network, including but not limited to, cloud computing systems. Modern neural networks are non-linear statistical data modeling tools. They are usually used to model complex relationships between inputs and outputs, or to identify patterns (e.g., states) in data (i.e., neural networks are non-linear statistical data modeling or decision-making tools). In general, program code utilizing neural networks can model complex relationships between inputs and outputs and identified patterns in data. Because of the speed and efficiency of neural networks, especially when parsing complex datasets, neural networks and deep learning provide solutions to many problems in multi-source processing, which program code, in embodiments of the present invention, can utilize in implementing a machine learning model, such as described herein.
Embodiments of the present invention include computer program products, computer systems, and computer-implemented methods, where program code executing on one or more processors performs a method, which includes obtaining training data for use in training a machine learning model, where the training data includes factual data related to a particular application, and obtaining, with reference to the training data, unlabeled counterfactual data for the particular application. The method further includes imputing one or more labels to the unlabeled counterfactual data using domain knowledge for the particular application, to obtain imputed counterfactual data, where the domain knowledge includes one or more application domain properties. Further, the method includes training the machine learning model using the training data and the imputed counterfactual data to facilitate generating machine learning model predictions for the particular application in accordance with the one or more application domain properties.
In one embodiment, the imputing of one or more labels to the unlabeled counterfactual data using the domain knowledge, and the training of the machine learning model using, in part, the imputed counterfactual data, facilitates integrating one or more monotonic-type application domain properties into the machine learning model.
In one or more implementations, the method further includes providing the trained machine learning model to an artificial intelligence explainability model to generate explanations for the imputed one or more labels to the unlabeled counterfactual data, and receiving an evaluation of the imputing obtained by checking the generated explanations for the imputing of the one or more labels to confirm presence of the one or more application domain properties.
In one embodiment, the method further includes using the trained machine learning model to provide a prediction for the particular application in accordance with the one or more monotonic-type application domain properties.
In one or more implementations, the training of the machine learning model using the training data and the imputed counterfactual data includes training the machine learning model using the training data, and updating the machine learning model using the imputed counterfactual data. In one embodiment, the method further includes iteratively re-imputing one or more labels to the counterfactual data, and updating the machine learning model using the re-imputed label(s) until convergence. In one or more embodiments, the updating the machine learning model using the imputed counterfactual data facilitates reducing prediction error of the machine learning model contrary to the one or more integrated monotonic-type application domain properties.
In one embodiment, the trained machine learning model simulates a randomized control trial in agreement with the one or more monotonic-type properties. Further, in one or more embodiments, the trained machine learning model for the particular application includes discrete actions and discrete outcomes.
One or more embodiments disclosed herein relates to machine learning models or algorithms that employ one or more artificial intelligence (AI) models or algorithms. AI models can include a trained machine learning model (e.g., models, such as neural network (NN), a convolution neural network (CNN), a recurrent neural network (RNN), a long-short-term memory (LSTM) based neural network, a gate recurrent unit (GRU) based recurrent neural network, a tree-based convolution neural network, a K-nearest neighbor (KNN) as a neural network, a self-attention network (e.g., a neural network that utilizes the attention mechanism as a basic building block; where self-attention networks have been shown to be effective for sequence modeling tasks, while having no recurrence or convolutions), a bi-directional long-short-term memory (BiLSTM), etc.). As known, an artificial neural network is an interconnected group of nodes or neurons.
One approach to machine learning is to learn an offline machine learning algorithm directly from observational data, and use the model for prediction. This is known as the “direct method”. The direct method functions typically without accounting for any inherent application domain property, such as one or more monotonic-type properties of a particular application. The direct method runs an online A/B test as a randomized trial that: requires carefully designed experiments and an interactive platform to support; can be costly (in terms of processing and memory) by lacking one or more application domain properties; the number of A/B tests is limited; and there is a relatively long turnaround time.
In accordance with one or more aspects disclosed herein, a model agnostic learning approach is disclosed to facilitate capturing one or more application domain attributes, such as one or more monotonicity properties. For instance, in one or more embodiments, a training dataset with labeled factual data and unlabeled counterfactual data can be augmented in a preprocessing step, where a known application domain property, such as a known monotonic-type property, can be used to impute labels for certain counterfactual data of the training dataset (based on the factual data available and the particular domain knowledge). In this way, the unlabeled counterfactual data can be leveraged by assigning labels to certain samples that are known to obey the desired attribute (e.g., monotonicity property), with the imputed counterfactual data being added to the training dataset used to train the model. In this manner, the attribute (e.g., monotonicity) can be encouraged in the learned function using the available application domain knowledge. The approach presented is flexible in that it can be used to train a variety of machine learning models, including, a direct method trained model. In addition to directly imputing certain counterfactual labels, other counterfactual labels can be further imputed by iteratively retraining the machine learning model or estimator, such as described in certain embodiments below.
In one or more embodiments, the counterfactual inference problem can be framed as a domain adaption problem, where the search domain is the factual data (i.e., observational data), and the target domain is a randomized trial on the same feature distribution. In one embodiment, a randomized trial is explicitly simulated by imputing labels (pseudo-labels) for the counterfactuals (e.g., the unobserved actions). The optimization process can be performed, in one embodiment, by directly imputing the pseudo-labels using one or more application domain properties and/or iteratively updating the pseudo-labels and using a machine learning model that is trained on both the factual and the counterfactual data with the desired property, or properties. This process can thus work in a preprocessing approach, and work in a self-supervised approach, referred to herein as a counterfactual self-training approach. With both imputing approaches, the one or more application domain properties are advantageously integrated into the machine learning model training.
For instance, in one or more aspects, in a self-supervised approach, rather than training a machine learning model (such as a raw classifier), with only existing factual data (i.e., observational data), unlabeled data can also be leveraged. For example, one or more known monotonic-type application domain properties of a particular application can be used to impute labels for certain counterfactual data obtained with reference to the training data set.
In one example embodiment, assume that there are 2 factual data in a data structure 320 where, for instance, entity A obtain or used a computing system with an attribute having value 2, denoted by “1” in the data structure, and entity B did not obtain or use a computing system with an attribute value 1 (denoted by “0”) in the data structure 320. The question marks in data structure 320 represent counterfactual outcomes, for which no factual data currently exists. For these unseen, counterfactual outcomes, pseudo-labels 330 can be imputed (either directly for certain outcomes, or indirectly, such as by a machine learning model), and then used to augment the factual data as part of the data used to update the machine learning model, or function. Depending on the implementation, for certain counterfactual outcomes, or even all counterfactual outcomes, pseudo-labels can be inserted into the revised training data structure 340, such as illustrated in
A variety of application domain properties, such as one or more monotonic-type properties, can be used to assign or impute labels, or pseudo-labels, to counterfactual data in the training data. For instance, where attribute value 1 is larger than attribute value 2, which is larger than attribute value 3, then, in the computing performance example, with entity A selecting the system with attribute value 2 performance, it can be imputed that all else being equal, entity A would also select the system with attribute value 1 performance. Similarly, since entity B did not select the system with attribute value 1 performance, it can be imputed that, all else being equal, entity B would not select the systems with attribute value 2 performance, or attribute value 3 performance. Thus, because of domain knowledge in the form of monotonicity of the demand function, labels for the unlabeled, counterfactual data can be assigned or imputed. As a result, instead of simply having two factual data samples, five factual data samples (for instance) are available as augmentation to train a machine learning model, such as a classifier. This can be a direct imputation process using the domain knowledge. The remaining unlabeled counterfactual data, that is, what would happen if entity A were presented with attribute value 3 performance, can further be obtained by iteratively retraining the estimator f(x,p,θ), or machine learning model.
As noted, in one embodiment, the data augmentation described herein, based on prior domain knowledge, can be implemented as a preprocessing step, one embodiment of which is depicted in
In a further embodiment, data augmentation such as described herein, based on prior domain knowledge, can be implemented in combination with a counterfactual self-training process such as discussed above in connection with
Referring to
Note in this regard that the pseudo-label imputation on counterfactual data based on domain knowledge is in combination with counterfactual self-training processing 530/550 to achieve the set of imputed counterfactual data 560 in the embodiment of
Self-training has been used in unsupervised domain adaptation (UDA) and semi-supervised learning (SSL), and has achieved success. The self-training algorithm works in an iterative fashion: first, after training a classifier f (x, p) on a source dataset, pseudo-labels are created by the best guess of f, along with the imputed labels based on application domain knowledge. Next, the model is trained on a target dataset, and the trained model is used to generate new pseudo-labels. This is illustrated in (x)π(p|x), where x represents an abstract space and
(x) is a probability distribution on x. The target domain is a randomized trial on the same feature distribution to ensure a uniformly good approximation on all actions. The goal is to transfer observational data from the source domain to a simulated pseudorandomized trial via self-training. To accomplish this, first an initial classifier f0(x, p) is trained on observational data, then pseudo-labels are imputed on all unseen actions from the observation data (and the imputing based on domain knowledge) with {circumflex over (r)}i,p˜f(xi, p). The model is then updated by training with the following objective:
In one embodiment, the first term src in Eq. 1 corresponds to the loss used in direct method, defined over the factual data alone. The second term refers to the loss defined over the imputed counterfactual data. In other words, in order to obtain a good model across all actions, the pseudo-population induced from imputation that represents a simulated randomized trial is utilized. The model is iteratively trained and pseudo-labels are imputed until the model converges.
As noted, in one embodiment, the iterative training is further with reference to pseudo-labeled counterfactual data based on domain knowledge 540. For instance, where an imputed label by the self-training agrees with the domain knowledge imputed label, it is retained, whereas if not, then it is re-imputed in a next iteration of the model update until the model converges.
In one embodiment, the output from the model update 530 is the pseudo-label imputation on counterfactuals
which provides the imputed counterfactual data
In block 570, as long as the iteration is valid, the process proceeds to again update the model 530. Otherwise, the process proceeds to the learnt function processing (f (x, p, θ)) 580 and the simulated randomized trial processing
By way of example,
Further details of one embodiment of facilitating processing within a computing environment, as it relates to one or more aspects of the present invention, are described with reference to
Referring to
In one embodiment, imputing one or more labels to the unlabeled counterfactual data using the domain knowledge, and training the machine learning model using, in part, the imputed counterfactual data, facilitates integrating one or more monotonic-type properties into the machine learning model 708.
In one embodiment, the method further includes providing the trained machine learning model to an artificial intelligence explainability model to generate explanations for the imputed one or more labels to the unlabeled counterfactual data 710, and receiving an evaluation of the imputing obtained by checking the generated explanations for the imputing of the one or more labels to confirm presence of one or more application domain properties 712.
As illustrated in
In one embodiment, training the machine learning model using the training data and the imputed counterfactual data includes training the machine learning model using the training data 716, and updating the machine learning model using the imputed counterfactual data 718.
In one or more implementations, the method further includes iteratively re-imputing one or more labels to the counterfactual data 720, and updating the machine learning model using the re-imputed label(s) until convergence 722.
In one embodiment, the updating of the machine learning model using the imputed counterfactual data facilitates reducing any prediction error of the machine learning model contrary to the one or more monotonic-type properties 724.
In one embodiment, the trained machine learning model simulates a randomized control trial in agreement with the one or more monotonic-type properties 726. In one or more implementations, the trained machine learning model for the particular application includes discrete actions and discrete outcomes 728.
Other variations and embodiments are possible.
Imputing counterfactual data (in association with machine learning model training) in accordance with one or more aspects of the present invention can be incorporated and used in many computing environments. One example computing environment is described with reference to
As shown in
Bus 811 is, for instance, a memory or a cache coherence bus, and bus 810 represents, e.g., one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include the Industry Standard Architecture (ISA), the Micro Channel Architecture (MCA), the Enhanced ISA (EISA), the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI).
As examples, one or more special purpose processors (e.g., adjunct processors) can be separate from but coupled to one or more general purpose processors and/or can be embedded within one or more general purpose processors. May variations are possible.
Memory 806 can include, for instance, a cache 812, such as a shared cache, which may be coupled to local caches 814 of processors 804 via, for instance, one or more buses 810. Further, memory 806 can include one or more programs or applications 816, at least one operating system 818, one or more computer readable program instructions 820 and one or more machine learning agents 822 training one or more machine learning models 821. Computer readable program instructions 820 and/or machine learning agent(s) 822 can be configured to carry out, or facilitate, functions of embodiments of aspects of the invention.
Computer system 802 can communicate via, e.g., I/O interfaces 808 with one or more external devices 830, such as a user terminal, a tape drive, a pointing device, a display, and one or more data storage devices 834, etc. A data storage device 834 can store one or more programs 836, one or more computer readable program instructions 838, and/or data, etc. The computer readable program instructions can be configured to carry out functions of embodiments of aspects of the invention.
Computer system 802 can also communicate via, e.g., I/O interfaces 808 with a network interface 832, which enables computer system 802 to communicate with one or more networks, such as a local area network (LAN), a general wide area network (WAN), and/or a public network (e.g., the Internet), providing communication with other computing devices or systems.
Computer system 802 can include and/or be coupled to removable/non-removable, volatile/non-volatile computer system storage media. For example, it can include and/or be coupled to a non-removable, non-volatile magnetic media (typically called a “hard drive”), a magnetic disk drive for reading from and writing to a removable, non-volatile magnetic disk (e.g., a “floppy disk”), and/or an optical disk drive for reading from or writing to a removable, non-volatile optical disk, such as a CD-ROM, DVD-ROM or other optical media. It should be understood that other hardware and/or software components could be used in conjunction with computer system 802. Examples, include, but are not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archival storage systems, etc.
Computer system 802 can be operational with numerous other general-purpose or special purpose computing system environments or configurations. Examples of well-known computing systems, environments, and/or configurations that may be suitable for use with computer system 802 include, but are not limited to, personal computer (PC) systems, server computer systems, thin clients, thick clients, handheld or laptop devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputer systems, mainframe computer systems, and distributed cloud computing environments that include any of the above systems or devices, and the like.
Another embodiment of a computing environment which can incorporate and use one or more aspects of the present invention is described with reference to
Native central processing unit 912 includes one or more native registers 920, such as one or more general purpose registers and/or one or more special purpose registers used during processing within the environment. These registers include information that represents the state of the environment at any particular point in time.
Moreover, native central processing unit 912 executes instructions and code that are stored in memory 914. In one particular example, the central processing unit executes emulator code 922 stored in memory 914. This code enables the computing environment configured in one architecture to emulate another architecture. For instance, emulator code 922 allows machines based on architectures other than the z/Architecture hardware architecture, such as PowerPC processors, HP Superdome servers or others, to emulate the z/Architecture hardware architecture and to execute software and instructions developed based on the z/Architecture hardware architecture.
Further details relating to emulator code 922 are described with reference to
Further, emulator code 922 includes an emulation control routine 940 to cause the native instructions to be executed. Emulation control routine 940 may cause native CPU 912 to execute a routine of native instructions that emulate one or more previously obtained guest instructions and, at the conclusion of such execution, return control to the instruction fetch routine to emulate the obtaining of the next guest instruction or a group of guest instructions. Execution of the native instructions 936 may include loading data into a register from memory 914; storing data back to memory from a register; or performing some type of arithmetic or logic operation, as determined by the translation routine.
Each routine is, for instance, implemented in software, which is stored in memory and executed by native central processing unit 912. In other examples, one or more of the routines or operations are implemented in firmware, hardware, software or some combination thereof. The registers of the emulated processor may be emulated using registers 920 of the native CPU or by using locations in memory 914. In embodiments, guest instructions 930, native instructions 936 and emulator code 922 may reside in the same memory or may be disbursed among different memory devices.
Further, in one embodiment, computing environment 910 includes one or more inference accelerators 915 coupled to memory 914. The one or more accelerators are defined in one architecture and are configured to emulate another architecture. For example, an accelerator obtains guest commands of the architecture being emulated, translates the guest commands into native commands of the one architecture and executes the native commands.
The computing environments described above are only examples of computing environments that can be used. Other environments, including but not limited to, non-partitioned environments, partitioned environments, cloud environments and/or emulated environments, may be used; embodiments are not limited to any one environment. Although various examples of computing environments are described herein, one or more aspects of the present invention may be used with many types of environments. The computing environments provided herein are only examples.
Each computing environment is capable of being configured to include one or more aspects of the present invention. For instance, each may be configured for an inference acceleration facility, in accordance with one or more aspects of the present invention.
Although various embodiments are described herein, many variations and other embodiments are possible without departing from a spirit of aspects of the present invention. It should be noted that, unless otherwise inconsistent, each aspect or feature described herein, and variants thereof, may be combinable with any other aspect or feature.
One or more aspects may relate to cloud computing.
It is to be understood 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, embodiments of the present invention are capable of being implemented in conjunction with any other type of computing environment now known or later developed.
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, 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 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 datacenter).
Rapid elasticity: capabilities can be rapidly and elastically provisioned, 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 user accounts). Resource usage can be monitored, controlled, and reported, 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 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 e-mail). 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 user-specific application configuration settings.
Platform as a Service (PaaS): the capability provided to the consumer is 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 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 service oriented with a focus on statelessness, low coupling, modularity, and semantic interoperability. At the heart of cloud computing is an infrastructure that includes 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, 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 include 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 imputing counterfactual data in association with machine learning model training processing 96, such as disclosed herein.
Aspects of the present invention can be a system, a method, and/or a computer program product at any possible technical detail level of integration. The computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present invention.
The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of the computer readable storage medium includes the following: 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), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.
Computer readable program instructions described herein can be downloaded to respective computing/processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and/or a wireless network. The network may comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. A network adapter card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing/processing device.
Computer readable program instructions for carrying out operations of the present invention may be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, configuration data for integrated circuitry, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++, or the like, and procedural programming languages, such as the “C” programming language or similar programming languages. The computer readable program instructions 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). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), or programmable logic arrays (PLA) may execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present invention.
Aspects of the present invention are described herein 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 readable program instructions.
These computer readable program instructions can be provided to a processor of a 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 readable program instructions can also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including instructions which implement aspects of the function/act specified in the flowchart and/or block diagram block or blocks.
The computer readable program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process, such that the instructions which execute on the computer, other programmable apparatus, or other device implement 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 blocks may occur out of the order noted in the Figures. For example, two blocks shown in succession may, in fact, be accomplished as one step, executed concurrently, substantially concurrently, in a partially or wholly temporally overlapping manner, 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.
In addition to the above, one or more aspects may be provided, offered, deployed, managed, serviced, etc. by a service provider who offers management of customer environments. For instance, the service provider can create, maintain, support, etc. computer code and/or a computer infrastructure that performs one or more aspects for one or more customers. In return, the service provider may receive payment from the customer under a subscription and/or fee agreement, as examples. Additionally or alternatively, the service provider may receive payment from the sale of advertising content to one or more third parties.
In one aspect, an application may be deployed for performing one or more embodiments. As one example, the deploying of an application comprises providing computer infrastructure operable to perform one or more embodiments.
As a further aspect, a computing infrastructure may be deployed comprising integrating computer readable code into a computing system, in which the code in combination with the computing system is capable of performing one or more embodiments.
As yet a further aspect, a process for integrating computing infrastructure comprising integrating computer readable code into a computer system may be provided. The computer system comprises a computer readable medium, in which the computer medium comprises one or more embodiments. The code in combination with the computer system is capable of performing one or more embodiments.
Although various embodiments are described above, these are only examples. For example, computing environments of other architectures can be used to incorporate and use one or more embodiments. Further, different instructions, commands or operations may be used. Additionally, different types of indications or tags may be specified. Many variations are possible.
Various aspects are described herein. Further, many variations are possible without departing from a spirit of aspects of the present invention. It should be noted that, unless otherwise inconsistent, each aspect or feature described herein, and variants thereof, may be combinable with any other aspect or feature.
Further, other types of computing environments can benefit and be used. As an example, a data processing system suitable for storing and/or executing program code is usable that includes at least two processors coupled directly or indirectly to memory elements through a system bus. The memory elements include, for instance, local memory employed during actual execution of the program code, bulk storage, and cache memory which provide temporary storage of at least some program code in order to reduce the number of times code must be retrieved from bulk storage during execution.
Input/output or I/O devices (including, but not limited to, keyboards, displays, pointing devices, DASD, tape, CDs, DVDs, thumb drives and other memory media, etc.) can be coupled to the system either directly or through intervening I/O controllers. Network adapters may also be coupled to the system to enable the data processing system to become coupled to other data processing systems or remote printers or storage devices through intervening private or public networks. Modems, cable modems, and Ethernet cards are just a few of the available types of network adapters.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting. 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, if any, 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 one or more embodiments has been presented for purposes of illustration and description but is not intended to be exhaustive or limited to in the form disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiment was chosen and described in order to best explain various aspects and the practical application, and to enable others of ordinary skill in the art to understand various embodiments with various modifications as are suited to the particular use contemplated.
