US20220114451A1

US20220114451A1 - Methods and apparatus for data enhanced automated model generation

Info

Publication number: US20220114451A1
Application number: US17/559,730
Authority: US
Inventors: Chaunté W. Lacewell; Juan Pablo Muñoz; Rajesh Poornachandran; Nilesh Jain; Anahita Bhiwandiwalla; Eriko Nurvitadhi; Abhijit Davare
Original assignee: Intel Corp
Current assignee: Intel Corp
Priority date: 2021-07-16
Filing date: 2021-12-22
Publication date: 2022-04-14
Also published as: US20220113978A1; CN115617502A; CN115617736A; WO2023287563A1

Abstract

Methods, apparatus, systems, and articles of manufacture for data enhanced automated model generation are disclosed. Example instructions, when executed, cause at least one processor to access a request to generate a machine learning model to perform a selected task, generate task knowledge based on a previously generated machine learning model, create a search space based on the task knowledge, and generate a machine learning model using neural architecture search, the neural architecture search beginning based on the search space.

Description

RELATED APPLICATIONS

This application claims priority to U.S. Provisional Patent Application No. 63/222,938, which was filed on Jul. 16, 2021. U.S. Provisional Patent Application No. 63/222,938 is hereby incorporated by reference in its entirety.

FIELD OF THE DISCLOSURE

This disclosure relates generally to machine learning and, more particularly, to methods and apparatus for data enhanced automated model generation.

BACKGROUND

Machine learning is an important enabling technology for the revolution currently underway in artificial intelligence, driving truly remarkable advances in fields such as object detection, image classification, speech recognition, natural language processing, and many more. Models are created using machine learning that, when utilized, enable an output to be generated based on an input. Neural architecture search enables various architectures to be searched when creating a machine learning model.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram of an example system implemented in accordance with the teachings of this disclosure for data enhanced automated model generation.

FIG. 2 is a block diagram of an example process flow utilizing the example system of FIG. 1.

FIG. 3 is a flowchart representative of example machine readable instructions and/or example operations that may be executed by example processor circuitry to implement the example knowledge builder circuitry and the example model builder circuitry of FIG. 1.

FIG. 4 is a flowchart representative of example machine readable instructions and/or example operations that may be executed by example processor circuitry to implement the example target hardware of FIG. 1.

FIG. 5 is a block diagram of an example processing platform including processor circuitry structured to execute the example machine readable instructions and/or the example operations of FIG. 3 to implement the example knowledge builder circuitry and the example model builder circuitry of FIG. 2.

FIG. 6 is a block diagram of an example implementation of the processor circuitry of FIG. 4.

FIG. 7 is a block diagram of another example implementation of the processor circuitry of FIG. 4.

FIG. 8 is a block diagram of an example software distribution platform (e.g., one or more servers) to distribute software (e.g., software corresponding to the example machine readable instructions of FIGS. 3 and/or 4) to client devices associated with end users and/or consumers (e.g., for license, sale, and/or use), retailers (e.g., for sale, re-sale, license, and/or sub-license), and/or original equipment manufacturers (OEMs) (e.g., for inclusion in products to be distributed to, for example, retailers and/or to other end users such as direct buy customers).

In general, the same reference numbers will be used throughout the drawing(s) and accompanying written description to refer to the same or like parts. The figures are not to scale.
As used in this patent, stating that any part (e.g., a layer, film, area, region, or plate) is in any way on (e.g., positioned on, located on, disposed on, or formed on, etc.) another part, indicates that the referenced part is either in contact with the other part, or that the referenced part is above the other part with one or more intermediate part(s) located therebetween.
Unless specifically stated otherwise, descriptors such as “first,” “second,” “third,” etc., are used herein without imputing or otherwise indicating any meaning of priority, physical order, arrangement in a list, and/or ordering in any way, but are merely used as labels and/or arbitrary names to distinguish elements for ease of understanding the disclosed examples. In some examples, the descriptor “first” may be used to refer to an element in the detailed description, while the same element may be referred to in a claim with a different descriptor such as “second” or “third.” In such instances, it should be understood that such descriptors are used merely for identifying those elements distinctly that might, for example, otherwise share a same name.
As used herein “substantially real time” refers to occurrence in a near instantaneous manner recognizing there may be real world delays for computing time, transmission, etc. Thus, unless otherwise specified, “substantially real time” refers to real time+/−1 second.
As used herein, the phrase “in communication,” including variations thereof, encompasses direct communication and/or indirect communication through one or more intermediary components, and does not require direct physical (e.g., wired) communication and/or constant communication, but rather additionally includes selective communication at periodic intervals, scheduled intervals, aperiodic intervals, and/or one-time events.
As used herein, “processor circuitry” is defined to include (i) one or more special purpose electrical circuits structured to perform specific operation(s) and including one or more semiconductor-based logic devices (e.g., electrical hardware implemented by one or more transistors), and/or (ii) one or more general purpose semiconductor-based electrical circuits programmed with instructions to perform specific operations and including one or more semiconductor-based logic devices (e.g., electrical hardware implemented by one or more transistors). Examples of processor circuitry include programmed microprocessors, Field Programmable Gate Arrays (FPGAs) that may instantiate instructions, Central Processor Units (CPUs), Graphics Processor Units (GPUs), Digital Signal Processors (DSPs), XPUs, or microcontrollers and integrated circuits such as Application Specific Integrated Circuits (ASICs). For example, an XPU may be implemented by a heterogeneous computing system including multiple types of processor circuitry (e.g., one or more FPGAs, one or more CPUs, one or more GPUs, one or more DSPs, etc., and/or a combination thereof) and application programming interface(s) (API(s)) that may assign computing task(s) to whichever one(s) of the multiple types of the processing circuitry is/are best suited to execute the computing task(s).

DETAILED DESCRIPTION

Neural Architecture Search (NAS) is an approach for exploring different machine learning algorithms for solving machine learning tasks. NAS algorithms take significant amount resources (e.g., compute resources, temporal resources, energy resources, etc.) to identify acceptable architectures. Most of these resources are expended by examining non-optimal architecture configurations during an exploration stage. Existing NAS algorithms do not provide clear explanations of the decisions for selecting a particular architecture, and such algorithms do not benefit from collected data regarding previous findings (e.g., sequence of operations, FLOPs, etc.) or target hardware capabilities. This information is typically discarded and does not benefit future applications of the NAS algorithm.
Due to the complexity of the task, NAS solutions tend to forget any insights from one run to the next. The initial conditions/configurations in previous solutions are independent of any other configurations used previously.
Existing NAS approaches do not reuse prior execution data related to models identified via NAS. That is, existing approaches do not benefit from collected knowledge about the task that the model will perform (e.g., detection, segmentation, etc.). When performing NAS, existing approaches start from scratch every time, when looking for better models. Many existing NAS approaches also require significant reconfiguration when moving to different tasks, and such approaches do not generalize the neural network architecture search process.
Example approaches disclosed herein analyze state-of-the-art and emerging workloads and collect historical information about the models including performance, sequence of operations, size, floating point operations per second (FLOPS), etc. for each operation.
In examples disclosed herein, a user provides a task (object recognition, segmentation, etc.) and objective (accuracy, latency, mix, etc.), and the NAS system selects starting hyperparameters/configuration information which include the best configuration for the task, objective, and, in some examples, the target hardware on which the model is to be executed.
Collected execution and/or performance information provides insights and guides the initial conditions on the search for an architecture that satisfies the requirements. The system also collects target hardware information, making the system hardware-aware and allowing the system to refine for the specific target hardware(s). For example, the system can avoid dilated 7×7 convolution kernels if kernel does not perform well (e.g., latency on the selected target hardware exceeds a threshold amount of latency).
Example approaches disclosed herein provide the user with the generated model and the reasoning behind the choices made when selecting operations. The decisions are based on the collected historical data and the task knowledge obtained from the knowledge builder (KB). Providing the reasoning for decisions can result in insights for future HW improvements (e.g., optimize specific kernels, memory BW, etc.)
FIG. 1 is a block diagram of an example system implemented in accordance with the teachings of this disclosure for data enhanced automated model generation. The example system 100 of FIG. 1 includes knowledge builder circuitry 105 that receives a user input 110, and model builder circuitry 115 that builds and provides a model to target hardware 120.
The example system of FIG. 1 presents an end-to-end solution that receives information from the user (objective, task, target HW), analyzes this information using a knowledge base and builds suggestions for the search space and initial configuration for the NAS approach. The approach is agnostic to the NAS approach to be used, enabling a user to decide on the state-of-the-art approach that will receive the suggested configuration.
The example user input 110 includes information including, for example, an objective of a machine learning model, a task to be performed by the machine learning model, and, optionally, one or more characteristics of a target hardware on which the machine learning model is to be executed. The task (object recognition, segmentation, etc.) will include input layer requirements, output layer requirements, and data requirements. The system of FIG. 1 is flexible enough that the user can provide information used to influence the model generation (e.g., by specifying whether the current task is similar to another task, and/or by specifying additional layers (not yet in the knowledge base, or associated with a different task) to include in the search space).
The knowledge builder circuitry 105 of FIG. 1 may be instantiated (e.g., creating an instance of, bring into being for any length of time, materialize, implement, etc.) by processor circuitry such as a central processing unit executing instructions. Additionally or alternatively, the knowledge builder circuitry 105 of FIG. 1 may be instantiated (e.g., creating an instance of, bring into being for any length of time, materialize, implement, etc.) by an ASIC or an FPGA structured to perform operations corresponding to the instructions. It should be understood that some or all of the circuitry of FIG. 1 may, thus, be instantiated at the same or different times (and/or by different hardware circuitry). Some or all of the circuitry may be instantiated, for example, in one or more threads executing concurrently on hardware and/or in series on hardware. Moreover, in some examples, some or all of the circuitry of FIG. 1 may be implemented by one or more virtual machines and/or containers executing on the microprocessor.
The example knowledge builder circuitry 105 of the illustrated example of FIG. 1 includes request accessor circuitry 130, hardware data orchestration circuitry 135, task data orchestration circuitry 140, and a knowledge datastore 145. The example knowledge builder circuitry 105 archives information for models and hardware into the knowledge datastore 145. If the hardware is not known in the knowledge datastore 145, the user is able to cause the system to execute on the target hardware 120 to extract performance metrics. A report of such performance metrics is obtained and added to the knowledge datastore 145 to build task knowledge. If the task is not in the knowledge datastore 145, the task data orchestration circuitry 140 creates task knowledge for the new tasks. FIG. 2 illustrates the process for creating or updating the knowledge datastore 145.
In examples disclosed herein, the knowledge datastore 145 of the knowledge builder circuitry 105 can be pre-populated with state-of-the-art (SOTA) or custom models and hardware configurations. In addition, the knowledge datastore 145 can be updated at any time based on, for example, statistics collected by the target hardware 120. In examples disclosed herein, the knowledge datastore 145 separates the models by tasks. To build the task knowledge, model information is retrieved from the knowledge datastore 145 the specific task and features are extracted from the models. In cases of a new or custom task, similar tasks/models are retrieved based on the user input. These features include, but are not limited to, the framework used to train the model, the HW specs and any information for mapping model (latencies, etc.) including HW telemetry, the performance objective, sequence of operations, number of FLOPs, dataset used, number of layers, etc. These features are then ranked by hardware features, objective, etc. The extracted and ranked features are then considered task knowledge which is then archived in the knowledge datastore 145 for future use.
The example request accessor circuitry 130 of the illustrated example of FIG. 1 receives a request for generation of a model to perform a selected task. In examples disclosed herein, the user input 110 received by the request accessor circuitry 130 includes information including, for example, an objective of a machine learning model, a task to be performed by the machine learning model, and, in some examples, one or more characteristics of a target hardware on which the machine learning model is to be executed. The request may be formatted as, for example, a request received at a web server, a request formatted in a structured data format (e.g., a JavaScript object notation (JSON) format, an extensible markup language (XML) format, etc.). The example request accessor circuitry 130 accesses hardware data orchestration information via the hardware data orchestration circuitry 135 and task data orchestration information via the task data orchestration circuitry 140. The accessed information (if available) and the request are provided to the search space management circuitry 160 of the model builder circuitry 115.
In some examples, the apparatus includes means for accessing a request. For example, the means for accessing may be implemented by the request accessor circuitry 130. In some examples, the request accessor circuitry 130 may be instantiated by processor circuitry such as the example processor circuitry 512 of FIG. 5. For instance, the request accessor circuitry 130 may be instantiated by the example general purpose processor circuitry 600 of FIG. 6 executing machine executable instructions such as that implemented by at least block 310 of FIG. 3. In some examples, the request accessor circuitry 130 may be instantiated by hardware logic circuitry, which may be implemented by an ASIC or the FPGA circuitry 700 of FIG. 7 structured to perform operations corresponding to the machine readable instructions. Additionally or alternatively, the request accessor circuitry 130 may be instantiated by any other combination of hardware, software, and/or firmware. For example, the request accessor circuitry 130 may be implemented by at least one or more hardware circuits (e.g., processor circuitry, discrete and/or integrated analog and/or digital circuitry, an FPGA, an Application Specific Integrated Circuit (ASIC), a comparator, an operational-amplifier (op-amp), a logic circuit, etc.) structured to execute some or all of the machine readable instructions and/or to perform some or all of the operations corresponding to the machine readable instructions without executing software or firmware, but other structures are likewise appropriate.
The example hardware data orchestration circuitry 135 of the illustrated example of FIG. 1 determines whether any prior knowledge is present in the knowledge datastore 145 for the selected hardware (e.g., the selected hardware identified in a request accessed by the request accessor circuitry 130). If no prior knowledge is known for the selected hardware, the example hardware data orchestration circuitry 135 adds an identification of the selected hardware to the knowledge datastore 145. The identification of the hardware enables subsequent performance metrics associated with the selected hardware to be stored in the knowledge datastore 145 in an organized fashion. In some examples, the identification of the selected hardware may be omitted prior to model creation and may, instead, be performed when performance metrics are provided to the knowledge datastore by the execution performance statistic collection circuitry 185.
The example task data orchestration circuitry 140 of the illustrated example of FIG. 1 determines whether any task information is available for the selected task. If no prior knowledge is available for the selected task, the example task data orchestration circuitry 140 adds an identification of the selected task to the knowledge datastore 145. The identification of the selected task enables subsequent performance metrics associated with the selected task to be stored in the knowledge datastore 145 in an organized fashion. In some examples, the identification of the selected task may be omitted prior to model creation and may, instead, be performed when performance metrics are provided to the knowledge datastore by the execution performance statistic collection circuitry 185.
In some examples, the apparatus includes means for generating task knowledge. For example, the means for generating task knowledge may be implemented by the example task data orchestration circuitry 140. In some examples, the example task data orchestration circuitry 140 may be instantiated by processor circuitry such as the example processor circuitry 512 of FIG. 5. For instance, the example task data orchestration circuitry 140 may be instantiated by the example general purpose processor circuitry 600 of FIG. 6 executing machine executable instructions such as that implemented by at least blocks 320, 335, 325 of FIG. 3. In some examples, the example task data orchestration circuitry 140 may be instantiated by hardware logic circuitry, which may be implemented by an ASIC or the FPGA circuitry 700 of FIG. 7 structured to perform operations corresponding to the machine readable instructions. Additionally or alternatively, the example task data orchestration circuitry 140 may be instantiated by any other combination of hardware, software, and/or firmware. For example, the example task data orchestration circuitry 140 may be implemented by at least one or more hardware circuits (e.g., processor circuitry, discrete and/or integrated analog and/or digital circuitry, an FPGA, an Application Specific Integrated Circuit (ASIC), a comparator, an operational-amplifier (op-amp), a logic circuit, etc.) structured to execute some or all of the machine readable instructions and/or to perform some or all of the operations corresponding to the machine readable instructions without executing software or firmware, but other structures are likewise appropriate.
The example knowledge datastore 145 of the illustrated example of FIG. 1 is implemented by any memory, storage device and/or storage disc for storing data such as, for example, flash memory, magnetic media, optical media, solid state memory, hard drive(s), thumb drive(s), etc. Furthermore, the data stored in the example knowledge datastore 145 may be in any data format such as, for example, binary data, comma delimited data, tab delimited data, structured query language (SQL) structures, etc. While, in the illustrated example, the knowledge datastore 145 is illustrated as a single device, the example knowledge datastore 145 and/or any other data storage devices described herein may be implemented by any number and/or type(s) of memories. In the illustrated example of FIG. 1, the example knowledge datastore 145 stores hardware and/or task knowledge.
The model builder circuitry 115 of FIG. 1 may be instantiated (e.g., creating an instance of, bring into being for any length of time, materialize, implement, etc.) by processor circuitry such as a central processing unit executing instructions. Additionally or alternatively, the model builder circuitry 115 of FIG. 1 may be instantiated (e.g., creating an instance of, bring into being for any length of time, materialize, implement, etc.) by an ASIC or an FPGA structured to perform operations corresponding to the instructions. As noted above, it should be understood that some or all of the circuitry of FIG. 1 may, thus, be instantiated at the same or different times (and/or by different hardware circuitry). Some or all of the circuitry may be instantiated, for example, in one or more threads executing concurrently on hardware and/or in series on hardware. Moreover, in some examples, some or all of the circuitry of FIG. 1 may be implemented by one or more virtual machines and/or containers executing on the microprocessor.
The example model builder circuitry 115 of the illustrated example of FIG. 1 includes search space management circuitry 160, anchor point inserter circuitry 165, neural architecture search circuitry 170, and model outputter circuitry 175. The model builder circuitry 115 is responsible for extracting the insights in the knowledge datastore and executing neural architecture search to identify an optimal model. First, the example search space management circuitry 160 creates a search space. This search space includes the operations provided by the task knowledge from the knowledge datastore, variants of those operations, and additional layers if the user specifies. The neural architecture search circuitry 170 performs a search that is initiated with the configuration identified by the search space management circuitry 160 for the objective, task, HW, etc. Anchor points are inserted in the chosen NAS algorithm by the anchor point inserter circuitry 165 to capture the decisions made during this process. The task knowledge is incorporated in the training loop of the neural architecture search circuitry 170 to inform decisions and guide the search. During training, historical decisions, confidence levels, and the knowledge datastore-based recommendations obtained from the task knowledge are used to guide the neural architecture search.
In some examples, the apparatus includes means for creating a search space. For example, the means for creating may be implemented by the example search space management circuitry 160. In some examples, the example search space management circuitry 160 may be instantiated by processor circuitry such as the example processor circuitry 512 of FIG. 5. For instance, the example search space management circuitry 160 may be instantiated by the example general purpose processor circuitry 500 of FIG. 5 executing machine executable instructions such as that implemented by at least blocks 327, 340 of FIG. 3. In some examples, the example search space management circuitry 160 may be instantiated by hardware logic circuitry, which may be implemented by an ASIC or the FPGA circuitry 700 of FIG. 7 structured to perform operations corresponding to the machine readable instructions. Additionally or alternatively, the example search space management circuitry 160 may be instantiated by any other combination of hardware, software, and/or firmware. For example, the example search space management circuitry 160 may be implemented by at least one or more hardware circuits (e.g., processor circuitry, discrete and/or integrated analog and/or digital circuitry, an FPGA, an Application Specific Integrated Circuit (ASIC), a comparator, an operational-amplifier (op-amp), a logic circuit, etc.) structured to execute some or all of the machine readable instructions and/or to perform some or all of the operations corresponding to the machine readable instructions without executing software or firmware, but other structures are likewise appropriate.
In some examples, the apparatus includes means for generating a machine learning model. For example, the means for generating may be implemented by the example neural architecture search circuitry 170. In some examples, the example neural architecture search circuitry 170 may be instantiated by processor circuitry such as the example processor circuitry 512 of FIG. 5. For instance, the example neural architecture search circuitry 170 may be instantiated by the example general purpose processor circuitry 600 of FIG. 6 executing machine executable instructions such as that implemented by at least blocks 330, 350 of FIG. 3. In some examples, the example neural architecture search circuitry 170 may be instantiated by hardware logic circuitry, which may be implemented by an ASIC or the FPGA circuitry 700 of FIG. 7 structured to perform operations corresponding to the machine readable instructions. Additionally or alternatively, the example neural architecture search circuitry 170 may be instantiated by any other combination of hardware, software, and/or firmware. For example, the example neural architecture search circuitry 170 may be implemented by at least one or more hardware circuits (e.g., processor circuitry, discrete and/or integrated analog and/or digital circuitry, an FPGA, an Application Specific Integrated Circuit (ASIC), a comparator, an operational-amplifier (op-amp), a logic circuit, etc.) structured to execute some or all of the machine readable instructions and/or to perform some or all of the operations corresponding to the machine readable instructions without executing software or firmware, but other structures are likewise appropriate.
In some examples, the apparatus includes means for inserting. For example, the means for inserting may be implemented by the example anchor point inserter circuitry 165. In some examples, the example anchor point inserter circuitry 165 may be instantiated by processor circuitry such as the example processor circuitry 512 of FIG. 5. For instance, the example anchor point inserter circuitry 165 may be instantiated by the example general purpose processor circuitry 600 of FIG. 6 executing machine executable instructions such as that implemented by at least block 360 of FIG. 3. In some examples, the example anchor point inserter circuitry 165 may be instantiated by hardware logic circuitry, which may be implemented by an ASIC or the FPGA circuitry 700 of FIG. 7 structured to perform operations corresponding to the machine readable instructions. Additionally or alternatively, the example anchor point inserter circuitry 165 may be instantiated by any other combination of hardware, software, and/or firmware. For example, the example anchor point inserter circuitry 165 may be implemented by at least one or more hardware circuits (e.g., processor circuitry, discrete and/or integrated analog and/or digital circuitry, an FPGA, an Application Specific Integrated Circuit (ASIC), a comparator, an operational-amplifier (op-amp), a logic circuit, etc.) structured to execute some or all of the machine readable instructions and/or to perform some or all of the operations corresponding to the machine readable instructions without executing software or firmware, but other structures are likewise appropriate.
After generation of the model, the example model outputter circuitry 175 provides a model for execution. In some examples, the decisions and/or rationales selected during the neural architecture search are made available in association with the generated model.
The target hardware 120 of FIG. 1 may be instantiated (e.g., creating an instance of, bring into being for any length of time, materialize, implement, etc.) by processor circuitry such as a central processing unit executing instructions. Additionally or alternatively, the target hardware 120 of FIG. 1 may be instantiated (e.g., creating an instance of, bring into being for any length of time, materialize, implement, etc.) by an ASIC or an FPGA structured to perform operations corresponding to the instructions. As noted above, it should be understood that some or all of the circuitry of FIG. 1 may, thus, be instantiated at the same or different times (and/or by different hardware circuitry). Some or all of the circuitry may be instantiated, for example, in one or more threads executing concurrently on hardware and/or in series on hardware. Moreover, in some examples, some or all of the circuitry of FIG. 1 may be implemented by one or more virtual machines and/or containers executing on the microprocessor.
The example target hardware 120 of the illustrated example of FIG. 1 includes model execution circuitry 180 and execution performance statistic collection circuitry 185. The example model execution circuitry 180 of the illustrated example of FIG. 1 executes the model provided by the model outputter circuitry 175.
The example execution performance statistic collection circuitry 185 of the illustrated example of FIG. 1, during execution of the model by the model execution circuitry 180, collects model execution statistics using the inserted anchor points. The collected execution statistics are provided to the knowledge datastore 145. In examples disclosed herein, the collected execution statistics include information about the anchor points. Including information about the anchor points enables statistics specific to particular features to be utilized when generating task knowledge.
FIG. 2 is a block diagram of an example process flow utilizing the example system of FIG. 1. The example process begins when a user submits a request for generation of a model to perform a selected task. (Blocks 210). The requested model is generated using neural architecture search and prior knowledge of models associated with the selected task. (Block 220). The generated models are provided to the target hardware for execution and collection of performance statistics. (Blocks 230). Execution features are extracted from the models. (Block 240). The extracted features are ranked based on collected performance metrics. (Block 250). The extracted features and their associated performance metrics are added to the knowledge datastore 145. (Block 260). This added knowledge may then subsequently be used for future generation of models. (Block 220).
While an example manner of implementing the example knowledge builder circuitry 105 and/or the example model builder circuitry 115 is illustrated in FIG. 1, one or more of the elements, processes, and/or devices illustrated in FIG. 1 may be combined, divided, re-arranged, omitted, eliminated, and/or implemented in any other way. Further, the example request accessor circuitry 130, the example hardware data orchestration circuitry 135, the example task data orchestration circuitry 140, and/or more, generally, example knowledge builder circuitry 105 of FIG. 1, and/or the example search space management circuitry 160, the example anchor point inserter circuitry 165, the example neural architecture search circuitry 170, the example model outputter circuitry 175, and/or, more generally, the example model builder circuitry 115 of FIG. 1, may be implemented by hardware alone or by hardware in combination with software and/or firmware. Thus, for example, any of the example request accessor circuitry 130, the example hardware data orchestration circuitry 135, the example task data orchestration circuitry 140, and/or more, generally, example knowledge builder circuitry 105 of FIG. 1, and/or the example search space management circuitry 160, the example anchor point inserter circuitry 165, the example neural architecture search circuitry 170, the example model outputter circuitry 175, and/or, more generally, the example model builder circuitry 115 of FIG. 1, could be implemented by processor circuitry, analog circuit(s), digital circuit(s), logic circuit(s), programmable processor(s), programmable microcontroller(s), graphics processing unit(s) (GPU(s)), digital signal processor(s) (DSP(s)), application specific integrated circuit(s) (ASIC(s)), programmable logic device(s) (PLD(s)), and/or field programmable logic device(s) (FPLD(s)) such as Field Programmable Gate Arrays (FPGAs). Further still, the example request accessor circuitry 130, the example hardware data orchestration circuitry 135, the example task data orchestration circuitry 140, and/or more, generally, example knowledge builder circuitry 105 of FIG. 1, and/or the example search space management circuitry 160, the example anchor point inserter circuitry 165, the example neural architecture search circuitry 170, the example model outputter circuitry 175, and/or, more generally, the example model builder circuitry 115 of FIG. 1 may include one or more elements, processes, and/or devices in addition to, or instead of, those illustrated in FIG. 1, and/or may include more than one of any or all of the illustrated elements, processes and devices.
A flowchart representative of example hardware logic circuitry, machine readable instructions, hardware implemented state machines, and/or any combination thereof for implementing the knowledge builder circuitry 105 and/or the example model builder circuitry 115 of FIG. 1 is shown in FIG. 3. The machine readable instructions may be one or more executable programs or portion(s) of an executable program for execution by processor circuitry, such as the processor circuitry 512 shown in the example processor platform 500 discussed below in connection with FIG. 5 and/or the example processor circuitry discussed below in connection with FIGS. 5 and/or 6.
A flowchart representative of example hardware logic circuitry, machine readable instructions, hardware implemented state machines, and/or any combination thereof for implementing the target hardware 120 of FIG. 1 is shown in FIG. 4. The machine readable instructions may be one or more executable programs or portion(s) of an executable program for execution by processor circuitry, such as the processor circuitry 512 shown in the example processor platform 500 discussed below in connection with FIG. 5 and/or the example processor circuitry discussed below in connection with FIGS. 5 and/or 6.
The programs of FIGS. 3 and/or 4 may be embodied in software stored on one or more non-transitory computer readable storage media such as a compact disk (CD), a floppy disk, a hard disk drive (HDD), a solid-state drive (SSD), a digital versatile disk (DVD), a Blu-ray disk, a volatile memory (e.g., Random Access Memory (RAM) of any type, etc.), or a non-volatile memory (e.g., electrically erasable programmable read-only memory (EEPROM), FLASH memory, an HDD, an SSD, etc.) associated with processor circuitry located in one or more hardware devices, but the entire program and/or parts thereof could alternatively be executed by one or more hardware devices other than the processor circuitry and/or embodied in firmware or dedicated hardware. The machine readable instructions may be distributed across multiple hardware devices and/or executed by two or more hardware devices (e.g., a server and a client hardware device). For example, the client hardware device may be implemented by an endpoint client hardware device (e.g., a hardware device associated with a user) or an intermediate client hardware device (e.g., a radio access network (RAN)) gateway that may facilitate communication between a server and an endpoint client hardware device). Similarly, the non-transitory computer readable storage media may include one or more mediums located in one or more hardware devices. Further, although the example program is described with reference to the flowchart illustrated in FIG. 3, many other methods of implementing the example knowledge builder circuitry 105 and/or the example model builder circuitry 115 may alternatively be used. For example, the order of execution of the blocks may be changed, and/or some of the blocks described may be changed, eliminated, or combined. Additionally or alternatively, any or all of the blocks may be implemented by one or more hardware circuits (e.g., processor circuitry, discrete and/or integrated analog and/or digital circuitry, an FPGA, an ASIC, a comparator, an operational-amplifier (op-amp), a logic circuit, etc.) structured to perform the corresponding operation without executing software or firmware. The processor circuitry may be distributed in different network locations and/or local to one or more hardware devices (e.g., a single-core processor (e.g., a single core central processor unit (CPU)), a multi-core processor (e.g., a multi-core CPU), etc.) in a single machine, multiple processors distributed across multiple servers of a server rack, multiple processors distributed across one or more server racks, a CPU and/or a FPGA located in the same package (e.g., the same integrated circuit (IC) package or in two or more separate housings, etc.).
The machine readable instructions described herein may be stored in one or more of a compressed format, an encrypted format, a fragmented format, a compiled format, an executable format, a packaged format, etc. Machine readable instructions as described herein may be stored as data or a data structure (e.g., as portions of instructions, code, representations of code, etc.) that may be utilized to create, manufacture, and/or produce machine executable instructions. For example, the machine readable instructions may be fragmented and stored on one or more storage devices and/or computing devices (e.g., servers) located at the same or different locations of a network or collection of networks (e.g., in the cloud, in edge devices, etc.). The machine readable instructions may require one or more of installation, modification, adaptation, updating, combining, supplementing, configuring, decryption, decompression, unpacking, distribution, reassignment, compilation, etc., in order to make them directly readable, interpretable, and/or executable by a computing device and/or other machine. For example, the machine readable instructions may be stored in multiple parts, which are individually compressed, encrypted, and/or stored on separate computing devices, wherein the parts when decrypted, decompressed, and/or combined form a set of machine executable instructions that implement one or more operations that may together form a program such as that described herein.
In another example, the machine readable instructions may be stored in a state in which they may be read by processor circuitry, but require addition of a library (e.g., a dynamic link library (DLL)), a software development kit (SDK), an application programming interface (API), etc., in order to execute the machine readable instructions on a particular computing device or other device. In another example, the machine readable instructions may need to be configured (e.g., settings stored, data input, network addresses recorded, etc.) before the machine readable instructions and/or the corresponding program(s) can be executed in whole or in part. Thus, machine readable media, as used herein, may include machine readable instructions and/or program(s) regardless of the particular format or state of the machine readable instructions and/or program(s) when stored or otherwise at rest or in transit.
The machine readable instructions described herein can be represented by any past, present, or future instruction language, scripting language, programming language, etc. For example, the machine readable instructions may be represented using any of the following languages: C, C++, Java, C#, Perl, Python, JavaScript, HyperText Markup Language (HTML), Structured Query Language (SQL), Swift, etc.
As mentioned above, the example operations of FIGS. 3 and/or 4 may be implemented using executable instructions (e.g., computer and/or machine readable instructions) stored on one or more non-transitory computer and/or machine readable media such as optical storage devices, magnetic storage devices, an HDD, a flash memory, a read-only memory (ROM), a CD, a DVD, a cache, a RAM of any type, a register, and/or any other storage device or storage disk in which information is stored for any duration (e.g., for extended time periods, permanently, for brief instances, for temporarily buffering, and/or for caching of the information). As used herein, the terms non-transitory computer readable medium and non-transitory computer readable storage medium are expressly defined to include any type of computer readable storage device and/or storage disk and to exclude propagating signals and to exclude transmission media.
“Including” and “comprising” (and all forms and tenses thereof) are used herein to be open ended terms. Thus, whenever a claim employs any form of “include” or “comprise” (e.g., comprises, includes, comprising, including, having, etc.) as a preamble or within a claim recitation of any kind, it is to be understood that additional elements, terms, etc., may be present without falling outside the scope of the corresponding claim or recitation. As used herein, when the phrase “at least” is used as the transition term in, for example, a preamble of a claim, it is open-ended in the same manner as the term “comprising” and “including” are open ended. The term “and/or” when used, for example, in a form such as A, B, and/or C refers to any combination or subset of A, B, C such as (1) A alone, (2) B alone, (3) C alone, (4) A with B, (5) A with C, (6) B with C, or (7) A with B and with C. As used herein in the context of describing structures, components, items, objects and/or things, the phrase “at least one of A and B” is intended to refer to implementations including any of (1) at least one A, (2) at least one B, or (3) at least one A and at least one B. Similarly, as used herein in the context of describing structures, components, items, objects and/or things, the phrase “at least one of A or B” is intended to refer to implementations including any of (1) at least one A, (2) at least one B, or (3) at least one A and at least one B. As used herein in the context of describing the performance or execution of processes, instructions, actions, activities and/or steps, the phrase “at least one of A and B” is intended to refer to implementations including any of (1) at least one A, (2) at least one B, or (3) at least one A and at least one B. Similarly, as used herein in the context of describing the performance or execution of processes, instructions, actions, activities and/or steps, the phrase “at least one of A or B” is intended to refer to implementations including any of (1) at least one A, (2) at least one B, or (3) at least one A and at least one B.
As used herein, singular references (e.g., “a”, “an”, “first”, “second”, etc.) do not exclude a plurality. The term “a” or “an” object, as used herein, refers to one or more of that object. The terms “a” (or “an”), “one or more”, and “at least one” are used interchangeably herein. Furthermore, although individually listed, a plurality of means, elements or method actions may be implemented by, e.g., the same entity or object. Additionally, although individual features may be included in different examples or claims, these may possibly be combined, and the inclusion in different examples or claims does not imply that a combination of features is not feasible and/or advantageous.
FIG. 3 is a flowchart representative of example machine readable instructions and/or example operations 300 that may be executed and/or instantiated by processor circuitry to implement the example knowledge builder circuitry and the example model builder circuitry of FIG. 1. The machine readable instructions and/or the operations 300 of FIG. 3 begin at block 310, at which the request accessor circuitry 130 receives a request for generation of a model to perform a selected task. (Block 310). In examples disclosed herein, the user input 110 received by the request accessor circuitry 130 includes information including, for example, an objective of a machine learning model, a task to be performed by the machine learning model, and, in some examples, one or more characteristics of a target hardware on which the machine learning model is to be executed. The request may be formatted as, for example, a request received at a web server, a request formatted in a structured data format (e.g., a JavaScript object notation (JSON) format, an extensible markup language (XML) format, etc.). The example request accessor circuitry 130 accesses hardware data orchestration information via the hardware data orchestration circuitry 135 and task data orchestration information via the task data orchestration circuitry 140. The accessed information (if available) and the request are provided to the search space management circuitry 160 of the model builder circuitry 115.
The example hardware data orchestration circuitry 135 determines whether any prior knowledge is present in the knowledge datastore 145 for the selected hardware. (Block 312). If no prior knowledge is known for the selected hardware (e.g., block 312 returns a result of NO), the example hardware data orchestration circuitry 135 adds an identification of the selected hardware to the knowledge datastore 145. (Block 314). The identification of the hardware enables subsequent performance metrics associated with the selected hardware to be stored in the knowledge datastore 145 in an organized fashion. In some examples, the identification of the selected hardware may be omitted prior to model creation and may, instead, be performed when performance metrics are provided to the knowledge datastore by the execution performance statistic collection circuitry 185.
The example task data orchestration circuitry 140 determines whether any task information is available for the selected task. (Block 320). If no prior knowledge is available for the selected task (e.g., block 320 returns a result of NO), the example task data orchestration circuitry 140 adds an identification of the selected task to the knowledge datastore 145. (Block 325). The identification of the selected task enables subsequent performance metrics associated with the selected task to be stored in the knowledge datastore 145 in an organized fashion. In some examples, the identification of the selected task may be omitted prior to model creation and may, instead, be performed when performance metrics are provided to the knowledge datastore by the execution performance statistic collection circuitry 185. The example search space management circuitry 160 creates a search space based on user selection of available building blocks or building blocks from existing state-of-the-art architecture(s) for the task. (Block 327). In this manner, the search space is created, but is not based on specific prior task knowledge (as is described in connection with block 340, below). In some examples, the ability to perform user selection of available building blocks (and/or whether to use state-of-the-art architecture(s) for the task) may be configurable by policy.
The example NAS search circuitry 170 performs neural architecture search to generate a model using the search space. (Block 330). In the illustrated example of FIG. 3, the NAS search circuitry 170 starts from an uninitialized state. That is, no prior knowledge of performance of various tasks and/or hardware on which the tasks are to be executed is used when performing the neural architecture search of block 330.
Returning to block 320, if the task data orchestration circuitry 140 determines that prior knowledge is present for the selected task (e.g., block 320 returns a result of YES), the example task data orchestration circuitry 140 builds task knowledge. (Block 335). To build the task knowledge, model information is retrieved by the task data orchestration circuitry 140 from the knowledge datastore 145 for the specific task and features are extracted from the models. In cases of a new or custom task, similar tasks/models are retrieved based on the user input. These features include, but are not limited to, the framework used to train the model, the hardware specification and/or any information for mapping model (latencies, etc.) including hardware telemetry, the performance objective, sequence of operations, number of FLOPs, dataset used, number of layers, etc. These features are then ranked by hardware, objective, etc. The respective features extracted and ranked from the model(s) is collectively identified as the task knowledge which is then used to create the search space. In some examples, such task knowledge is archived in the knowledge datastore 145 to allow for efficient retrieval should a same task be later requested.
The example search space management circuitry 160 creates a search space from the prior task knowledge. (Block 340). The search space may be created by, for example, ranking and selecting a prior architecture that had an acceptable level of performance on the target hardware (and/or hardware similar to the target hardware). In some examples, performance statistics stored in the knowledge datastore 145 associated with different architectures and tasks are compared to select an architecture meeting a threshold performance statistic. In some examples, the performance statistic upon which the selection is based may be dependent upon the user input 110 which may indicate, for example, whether power consumption statistics are to be prioritized over processing speed statistics.
In some examples, the selection of the prioritization (e.g., prioritization of functionality, performance, power optimization, etc.) may be guided by a policy. For example, a policy may be provided by a policy-providing entity to control behavior of the training operations and/or search space management. In some examples, the policy controls other details about the creation and/or training of the model including, for example, different levels of neural network sparsity (e.g., 50%, 90%, etc.), different levels of precision (e.g., thirty-two bit floating point values, sixteen-bit floating point values, eight bit integer values, etc.)
In some examples, the policy-providing entity may be a user of the system of FIG. 1. However, the policy-providing entity may be any other entity that guides functionality of the system of FIG. 1 including, for example, a system administrator, a manufacturer, a device provider, etc. In some examples, the policy-providing entity may be separate from the user. In this manner, the user is able to input requests for training and/or creation of a machine learning model, while allowing the parameters under which the training and/or creation of the machine learning model to be based on the policy created by the policy-providing entity.
In some examples the policy is provisioned to the system of FIG. 1 by the policy-providing entity via a platform Trusted Execution Environment (TEE). However, the policy may be provided to the system of FIG. 1 in any other manner.
The example NAS search circuitry 170 generates a model using neural architecture search, based on the search space created by the search space management circuitry 160. (Block 350). In this manner, the neural architecture search performed by the NAS search circuitry 170 at block 350 starts from an initialized state based on the prior task knowledge (e.g., starting from an architecture which previously met a performance threshold).
The example anchor point inserter circuitry 165 then inserts anchor points into the generated model. (Block 360). Anchor points provide locations at which performance statistics are to be measured by the execution performance statistic collection circuitry 185. Moreover, the anchor points provide locations by which additional information about the model and/or the objectives/tasks of the model may be captured. In examples disclosed herein, anchor points are inserted intermediate respective layers of the generated model. In some examples, anchor points are added to the model prior to the first layer and after the last layer of the model. In some other examples, anchor points are added adjacent (e.g., before and after) particular types of layers (e.g., a convolution layer).
The example model outputter circuitry 175 provides the generated model to the target hardware 120 for execution by the model execution circuitry 180. (Block 370). In examples disclosed herein, the model may first be stored at a storage location (e.g., a server) before being provided to the model execution circuitry 180. In some examples, the model execution circuitry 180 may retrieve the model from the storage location or directly from the model outputter circuitry 175. The process of the illustrated example of FIG. 3 then terminates, but by may be re-executed upon, for example, receipt of subsequent user input 110.
FIG. 4 is a flowchart representative of example machine readable instructions and/or example operations 400 that may be executed and/or instantiated by processor circuitry to implement the example target hardware 120 of FIG. 1. The machine readable instructions and/or the operations 400 of FIG. 4 begin at block 410, at which the model execution circuitry 180 begin execution of a model received from the model outputter circuitry 175. (Block 410). During execution of the model, the example execution performance statistic collection circuitry 185 collects model execution statistics using the inserted anchor points. (Block 420). The collected execution statistics are provided to the knowledge datastore 145. (Block 430). In examples disclosed herein, the collected execution statistics include information about the anchor points. Including information about the anchor points enables statistics specific to particular features to be utilized when generating task knowledge.
FIG. 5 is a block diagram of an example processor platform 500 structured to execute and/or instantiate the machine readable instructions and/or the operations of FIGS. 3 and/or 4 to implement the system 100 of FIG. 1. The processor platform 500 can be, for example, a server, a personal computer, a workstation, a self-learning machine (e.g., a neural network), a mobile device (e.g., a cell phone, a smart phone, a tablet such as an iPad™), a personal digital assistant (PDA), an Internet appliance, a DVD player, a CD player, a digital video recorder, a Blu-ray player, a gaming console, a personal video recorder, a set top box, a headset (e.g., an augmented reality (AR) headset, a virtual reality (VR) headset, etc.) or other wearable device, or any other type of computing device.
The processor platform 500 of the illustrated example includes processor circuitry 512. The processor circuitry 512 of the illustrated example is hardware. For example, the processor circuitry 512 can be implemented by one or more integrated circuits, logic circuits, FPGAs, microprocessors, CPUs, GPUs, DSPs, and/or microcontrollers from any desired family or manufacturer. The processor circuitry 512 may be implemented by one or more semiconductor based (e.g., silicon based) devices. In this example, the processor circuitry 512 implements the knowledge builder circuitry 105 and the model builder circuitry 115. In some examples, the knowledge builder circuitry 105 and the model builder circuitry 115 may be implemented on separate processor platforms.
The processor circuitry 512 of the illustrated example includes a local memory 513 (e.g., a cache, registers, etc.). The processor circuitry 512 of the illustrated example is in communication with a main memory including a volatile memory 514 and a non-volatile memory 516 by a bus 518. The volatile memory 514 may be implemented by Synchronous Dynamic Random Access Memory (SDRAM), Dynamic Random Access Memory (DRAM), RAMBUS® Dynamic Random Access Memory (RDRAM®), and/or any other type of RAM device. The non-volatile memory 516 may be implemented by flash memory and/or any other desired type of memory device. Access to the main memory 514, 516 of the illustrated example is controlled by a memory controller 517.
The processor platform 500 of the illustrated example also includes interface circuitry 520. The interface circuitry 520 may be implemented by hardware in accordance with any type of interface standard, such as an Ethernet interface, a universal serial bus (USB) interface, a Bluetooth® interface, a near field communication (NFC) interface, a Peripheral Component Interconnect (PCI) interface, and/or a Peripheral Component Interconnect Express (PCIe) interface.
In the illustrated example, one or more input devices 522 are connected to the interface circuitry 520. The input device(s) 522 permit(s) a user to enter data and/or commands into the processor circuitry 512. The input device(s) 522 can be implemented by, for example, an audio sensor, a microphone, a camera (still or video), a keyboard, a button, a mouse, a touchscreen, a track-pad, a trackball, an isopoint device, and/or a voice recognition system.
One or more output devices 524 are also connected to the interface circuitry 520 of the illustrated example. The output device(s) 524 can be implemented, for example, by display devices (e.g., a light emitting diode (LED), an organic light emitting diode (OLED), a liquid crystal display (LCD), a cathode ray tube (CRT) display, an in-place switching (IPS) display, a touchscreen, etc.), a tactile output device, a printer, and/or speaker. The interface circuitry 520 of the illustrated example, thus, typically includes a graphics driver card, a graphics driver chip, and/or graphics processor circuitry such as a GPU.
The interface circuitry 520 of the illustrated example also includes a communication device such as a transmitter, a receiver, a transceiver, a modem, a residential gateway, a wireless access point, and/or a network interface to facilitate exchange of data with external machines (e.g., computing devices of any kind) by a network 526. The communication can be by, for example, an Ethernet connection, a digital subscriber line (DSL) connection, a telephone line connection, a coaxial cable system, a satellite system, a line-of-site wireless system, a cellular telephone system, an optical connection, etc.
The processor platform 500 of the illustrated example also includes one or more mass storage devices 528 to store software and/or data. Examples of such mass storage devices 528 include magnetic storage devices, optical storage devices, floppy disk drives, HDDs, CDs, Blu-ray disk drives, redundant array of independent disks (RAID) systems, solid state storage devices such as flash memory devices and/or SSDs, and DVD drives.
The machine executable instructions 532, which may be implemented by the machine readable instructions of FIGS. 3 and/or 4, may be stored in the mass storage device 528, in the volatile memory 514, in the non-volatile memory 516, and/or on a removable non-transitory computer readable storage medium such as a CD or DVD.
FIG. 5 is a block diagram of an example implementation of the processor circuitry 512 of FIG. 5. In this example, the processor circuitry 512 of FIG. 5 is implemented by a general purpose microprocessor 600. The general purpose microprocessor circuitry 600 executes some or all of the machine readable instructions of the flowchart of FIG. 3 to effectively instantiate the knowledge builder circuitry 105 and/or the example model builder circuitry 115 of FIG. 1 as logic circuits to perform the operations corresponding to those machine readable instructions. In some such examples, the circuitry of FIG. 1 is instantiated by the hardware circuits of the microprocessor 600 in combination with the instructions. For example, the microprocessor 600 may implement multi-core hardware circuitry such as a CPU, a DSP, a GPU, an XPU, etc. Although it may include any number of example cores 602 (e.g., 1 core), the microprocessor 600 of this example is a multi-core semiconductor device including N cores. The cores 602 of the microprocessor 600 may operate independently or may cooperate to execute machine readable instructions. For example, machine code corresponding to a firmware program, an embedded software program, or a software program may be executed by one of the cores 602 or may be executed by multiple ones of the cores 602 at the same or different times. In some examples, the machine code corresponding to the firmware program, the embedded software program, or the software program is split into threads and executed in parallel by two or more of the cores 602. The software program may correspond to a portion or all of the machine readable instructions and/or operations represented by the flowchart of FIG. 3.
The cores 602 may communicate by a first example bus 604. In some examples, the first bus 604 may implement a communication bus to effectuate communication associated with one(s) of the cores 602. For example, the first bus 604 may implement at least one of an Inter-Integrated Circuit (I2C) bus, a Serial Peripheral Interface (SPI) bus, a PCI bus, or a PCIe bus. Additionally or alternatively, the first bus 604 may implement any other type of computing or electrical bus. The cores 602 may obtain data, instructions, and/or signals from one or more external devices by example interface circuitry 606. The cores 602 may output data, instructions, and/or signals to the one or more external devices by the interface circuitry 606. Although the cores 602 of this example include example local memory 620 (e.g., Level 1 (L1) cache that may be split into an L1 data cache and an L1 instruction cache), the microprocessor 600 also includes example shared memory 610 that may be shared by the cores (e.g., Level 2 (L2_cache)) for high-speed access to data and/or instructions. Data and/or instructions may be transferred (e.g., shared) by writing to and/or reading from the shared memory 610. The local memory 620 of each of the cores 602 and the shared memory 610 may be part of a hierarchy of storage devices including multiple levels of cache memory and the main memory (e.g., the main memory 514, 516 of FIG. 5). Typically, higher levels of memory in the hierarchy exhibit lower access time and have smaller storage capacity than lower levels of memory. Changes in the various levels of the cache hierarchy are managed (e.g., coordinated) by a cache coherency policy.
Each core 602 may be referred to as a CPU, DSP, GPU, etc., or any other type of hardware circuitry. Each core 602 includes control unit circuitry 614, arithmetic and logic (AL) circuitry (sometimes referred to as an ALU) 616, a plurality of registers 618, the L1 cache 620, and a second example bus 622. Other structures may be present. For example, each core 602 may include vector unit circuitry, single instruction multiple data (SIMD) unit circuitry, load/store unit (LSU) circuitry, branch/jump unit circuitry, floating-point unit (FPU) circuitry, etc. The control unit circuitry 614 includes semiconductor-based circuits structured to control (e.g., coordinate) data movement within the corresponding core 602. The AL circuitry 616 includes semiconductor-based circuits structured to perform one or more mathematic and/or logic operations on the data within the corresponding core 602. The AL circuitry 616 of some examples performs integer based operations. In other examples, the AL circuitry 616 also performs floating point operations. In yet other examples, the AL circuitry 616 may include first AL circuitry that performs integer based operations and second AL circuitry that performs floating point operations. In some examples, the AL circuitry 616 may be referred to as an Arithmetic Logic Unit (ALU). The registers 618 are semiconductor-based structures to store data and/or instructions such as results of one or more of the operations performed by the AL circuitry 616 of the corresponding core 602. For example, the registers 618 may include vector register(s), SIMD register(s), general purpose register(s), flag register(s), segment register(s), machine specific register(s), instruction pointer register(s), control register(s), debug register(s), memory management register(s), machine check register(s), etc. The registers 618 may be arranged in a bank as shown in FIG. 5. Alternatively, the registers 618 may be organized in any other arrangement, format, or structure including distributed throughout the core 602 to shorten access time. The second bus 622 may implement at least one of an I2C bus, a SPI bus, a PCI bus, or a PCIe bus
Each core 602 and/or, more generally, the microprocessor 600 may include additional and/or alternate structures to those shown and described above. For example, one or more clock circuits, one or more power supplies, one or more power gates, one or more cache home agents (CHAs), one or more converged/common mesh stops (CMSs), one or more shifters (e.g., barrel shifter(s)) and/or other circuitry may be present. The microprocessor 600 is a semiconductor device fabricated to include many transistors interconnected to implement the structures described above in one or more integrated circuits (ICs) contained in one or more packages. The processor circuitry may include and/or cooperate with one or more accelerators. In some examples, accelerators are implemented by logic circuitry to perform certain tasks more quickly and/or efficiently than can be done by a general purpose processor. Examples of accelerators include ASICs and FPGAs such as those discussed herein. A GPU or other programmable device can also be an accelerator. Accelerators may be on-board the processor circuitry, in the same chip package as the processor circuitry and/or in one or more separate packages from the processor circuitry.
FIG. 7 is a block diagram of another example implementation of the processor circuitry 512 of FIG. 5. In this example, the processor circuitry 512 is implemented by FPGA circuitry 700. The FPGA circuitry 700 can be used, for example, to perform operations that could otherwise be performed by the example microprocessor 600 of FIG. 6 executing corresponding machine readable instructions. However, once configured, the FPGA circuitry 700 instantiates the machine readable instructions in hardware and, thus, can often execute the operations faster than they could be performed by a general purpose microprocessor executing the corresponding software.
More specifically, in contrast to the microprocessor 600 of FIG. 6 described above (which is a general purpose device that may be programmed to execute some or all of the machine readable instructions represented by the flowchart of FIG. 3 but whose interconnections and logic circuitry are fixed once fabricated), the FPGA circuitry 700 of the example of FIG. 7 includes interconnections and logic circuitry that may be configured and/or interconnected in different ways after fabrication to instantiate, for example, some or all of the machine readable instructions represented by the flowchart of FIG. 3. In particular, the FPGA 700 may be thought of as an array of logic gates, interconnections, and switches. The switches can be programmed to change how the logic gates are interconnected by the interconnections, effectively forming one or more dedicated logic circuits (unless and until the FPGA circuitry 700 is reprogrammed). The configured logic circuits enable the logic gates to cooperate in different ways to perform different operations on data received by input circuitry. Those operations may correspond to some or all of the software represented by the flowchart of FIG. 3. As such, the FPGA circuitry 700 may be structured to effectively instantiate some or all of the machine readable instructions of the flowchart of FIG. 3 as dedicated logic circuits to perform the operations corresponding to those software instructions in a dedicated manner analogous to an ASIC. Therefore, the FPGA circuitry 700 may perform the operations corresponding to the some or all of the machine readable instructions of FIG. 3 faster than the general purpose microprocessor can execute the same.
In the example of FIG. 6, the FPGA circuitry 700 is structured to be programmed (and/or reprogrammed one or more times) by an end user by a hardware description language (HDL) such as Verilog. The FPGA circuitry 700 of FIG. 7, includes example input/output (I/O) circuitry 702 to obtain and/or output data to/from example configuration circuitry 704 and/or external hardware (e.g., external hardware circuitry) 706. For example, the configuration circuitry 704 may implement interface circuitry that may obtain machine readable instructions to configure the FPGA circuitry 700, or portion(s) thereof. In some such examples, the configuration circuitry 704 may obtain the machine readable instructions from a user, a machine (e.g., hardware circuitry (e.g., programmed or dedicated circuitry) that may implement an Artificial Intelligence/Machine Learning (AI/ML) model to generate the instructions), etc. In some examples, the external hardware 706 may implement the microprocessor 600 of FIG. 6. The FPGA circuitry 700 also includes an array of example logic gate circuitry 708, a plurality of example configurable interconnections 710, and example storage circuitry 712. The logic gate circuitry 708 and interconnections 710 are configurable to instantiate one or more operations that may correspond to at least some of the machine readable instructions of FIG. 3 and/or other desired operations. The logic gate circuitry 708 shown in FIG. 7 is fabricated in groups or blocks. Each block includes semiconductor-based electrical structures that may be configured into logic circuits. In some examples, the electrical structures include logic gates (e.g., And gates, Or gates, Nor gates, etc.) that provide basic building blocks for logic circuits. Electrically controllable switches (e.g., transistors) are present within each of the logic gate circuitry 708 to enable configuration of the electrical structures and/or the logic gates to form circuits to perform desired operations. The logic gate circuitry 708 may include other electrical structures such as look-up tables (LUTs), registers (e.g., flip-flops or latches), multiplexers, etc.
The interconnections 710 of the illustrated example are conductive pathways, traces, vias, or the like that may include electrically controllable switches (e.g., transistors) whose state can be changed by programming (e.g., using an HDL instruction language) to activate or deactivate one or more connections between one or more of the logic gate circuitry 708 to program desired logic circuits.
The storage circuitry 712 of the illustrated example is structured to store result(s) of the one or more of the operations performed by corresponding logic gates. The storage circuitry 712 may be implemented by registers or the like. In the illustrated example, the storage circuitry 712 is distributed amongst the logic gate circuitry 708 to facilitate access and increase execution speed.
The example FPGA circuitry 700 of FIG. 7 also includes example Dedicated Operations Circuitry 714. In this example, the Dedicated Operations Circuitry 714 includes special purpose circuitry 716 that may be invoked to implement commonly used functions to avoid the need to program those functions in the field. Examples of such special purpose circuitry 716 include memory (e.g., DRAM) controller circuitry, PCIe controller circuitry, clock circuitry, transceiver circuitry, memory, and multiplier-accumulator circuitry. Other types of special purpose circuitry may be present. In some examples, the FPGA circuitry 700 may also include example general purpose programmable circuitry 718 such as an example CPU 720 and/or an example DSP 722. Other general purpose programmable circuitry 718 may additionally or alternatively be present such as a GPU, an XPU, etc., that can be programmed to perform other operations.
Although FIGS. 6 and 7 illustrate two example implementations of the processor circuitry 512 of FIG. 5, many other approaches are contemplated. For example, as mentioned above, modern FPGA circuitry may include an on-board CPU, such as one or more of the example CPU 720 of FIG. 7. Therefore, the processor circuitry 512 of FIG. 5 may additionally be implemented by combining the example microprocessor 600 of FIG. 6 and the example FPGA circuitry 700 of FIG. 7. In some such hybrid examples, a first portion of the machine readable instructions represented by the flowchart of FIG. 3 may be executed by one or more of the cores 502 of FIG. 5, a second portion of the machine readable instructions represented by the flowchart of FIG. 3 may be executed by the FPGA circuitry 700 of FIG. 7, and/or a third portion of the machine readable instructions represented by the flowchart of FIG. 3 may be executed by an ASIC. It should be understood that some or all of the circuitry of FIG. 1 may, thus, be instantiated at the same or different times. Some or all of the circuitry may be instantiated, for example, in one or more threads executing concurrently and/or in series. Moreover, in some examples, some or all of the circuitry of FIG. 1 may be implemented within one or more virtual machines and/or containers executing on the microprocessor.
In some examples, the processor circuitry 512 of FIG. 5 may be in one or more packages. For example, the processor circuitry 600 of FIG. 6 and/or the FPGA circuitry 700 of FIG. 7 may be in one or more packages. In some examples, an XPU may be implemented by the processor circuitry 512 of FIG. 5, which may be in one or more packages. For example, the XPU may include a CPU in one package, a DSP in another package, a GPU in yet another package, and an FPGA in still yet another package.
A block diagram illustrating an example software distribution platform 805 to distribute software such as the example machine readable instructions 532 of FIG. 5 to hardware devices owned and/or operated by third parties is illustrated in FIG. 8. The example software distribution platform 805 may be implemented by any computer server, data facility, cloud service, etc., capable of storing and transmitting software to other computing devices. The third parties may be customers of the entity owning and/or operating the software distribution platform 805. For example, the entity that owns and/or operates the software distribution platform 805 may be a developer, a seller, and/or a licensor of software such as the example machine readable instructions 532 of FIG. 5. The third parties may be consumers, users, retailers, OEMs, etc., who purchase and/or license the software for use and/or re-sale and/or sub-licensing. In the illustrated example, the software distribution platform 805 includes one or more servers and one or more storage devices. The storage devices store the machine readable instructions 532, which may correspond to the example machine readable instructions 300, 400 of FIGS. 3 and/or 4, as described above. The one or more servers of the example software distribution platform 805 are in communication with a network 810, which may correspond to any one or more of the Internet and/or any of the example networks 526 described above. In some examples, the one or more servers are responsive to requests to transmit the software to a requesting party as part of a commercial transaction. Payment for the delivery, sale, and/or license of the software may be handled by the one or more servers of the software distribution platform and/or by a third party payment entity. The servers enable purchasers and/or licensors to download the machine readable instructions 532 from the software distribution platform 805. For example, the software, which may correspond to the example machine readable instructions 532 of FIG. 5, may be downloaded to the example processor platform 500, which is to execute the machine readable instructions 532 to implement the example knowledge builder circuitry 105, the example model builder circuitry 115, and/or the example target hardware 120 of FIG. 1. In some examples, one or more servers of the software distribution platform 805 periodically offer, transmit, and/or force updates to the software (e.g., the example machine readable instructions 532 of FIG. 5) to ensure improvements, patches, updates, etc., are distributed and applied to the software at the end user devices.
From the foregoing, it will be appreciated that example systems, methods, apparatus, and articles of manufacture have been disclosed that enable neural architecture search to be performed based on prior knowledge of models created to perform particular tasks. Disclosed systems, methods, apparatus, and articles of manufacture improve the efficiency of using a computing device by avoiding re-discovery of models that would otherwise be initially discovered by neural architecture search, but that do not function well for the intended task. By starting from based on prior knowledge, higher performing models can be identified more quickly. This reduces resource consumption not only on the target hardware (e.g., more efficient models can be developed), but also reduces resource consumption on systems that generate models (e.g., higher performing models can be discovered more quickly/efficiently). Disclosed systems, methods, apparatus, and articles of manufacture are accordingly directed to one or more improvement(s) in the operation of a machine such as a computer or other electronic and/or mechanical device.
Example methods, apparatus, systems, and articles of manufacture for data enhanced automated model generation are disclosed herein. Further examples and combinations thereof include the following:
Example 1 includes an apparatus for data enhanced automated model generation, the apparatus comprising interface circuitry to access a request to generate a machine learning model, and processor circuitry including one or more of at least one of a central processing unit, a graphic processing unit, or a digital signal processor, the at least one of the central processing unit, the graphic processing unit, or the digital signal processor having control circuitry to control data movement within the processor circuitry, arithmetic and logic circuitry to perform one or more first operations corresponding to instructions, and one or more registers to store a result of the one or more first operations, the instructions in the apparatus, a Field Programmable Gate Array (FPGA), the FPGA including logic gate circuitry, a plurality of configurable interconnections, and storage circuitry, the logic gate circuitry and interconnections to perform one or more second operations, the storage circuitry to store a result of the one or more second operations, or Application Specific Integrate Circuitry (ASIC) including logic gate circuitry to perform one or more third operations, the processor circuitry to perform at least one of the first operations, the second operations, or the third operations to instantiate task data orchestration circuitry to generate task knowledge based on a previously generated machine learning model, search space management circuitry to create a search space based on the task knowledge, and neural architecture search circuitry to generate the machine learning model using neural architecture search, the neural architecture search circuitry to begin an architecture search based on the search space.
Example 2 includes the apparatus of example 1, wherein the processor circuitry is to, during generation of the machine learning model, insert a plurality of anchor points into the machine learning model, the anchor points to be used for collection of a performance statistic concerning execution of the machine learning model.
Example 3 includes the apparatus of example 2, wherein the performance statistic includes at least one of power efficiency or energy efficiency.
Example 4 includes the apparatus of example 2, wherein the processor circuitry is further to collect the performance statistic based on the anchor points.
Example 5 includes the apparatus of example 4, wherein, to generate the task knowledge, the processor circuitry is further to rank features of the previously generated machine learning model.
Example 6 includes the apparatus of example 1, wherein to create the search space, the processor circuitry is to select a prior architecture based on performance of the prior architecture on a selected hardware.
Example 7 includes At least one non-transitory computer readable storage medium comprising instructions that, when executed, cause at least one processor to at least access a request to generate a machine learning model to perform a selected task, generate task knowledge based on a previously generated machine learning model, create a search space based on the task knowledge, and generate a machine learning model using neural architecture search, the neural architecture search beginning based on the search space.
Example 8 includes the at least one non-transitory computer readable storage medium of example 7, wherein the instructions, when executed, further cause the at least one processor to insert a plurality of anchor points into the machine learning model, the anchor points to be used when collecting a performance statistic concerning execution of the machine learning model.
Example 9 includes the at least one non-transitory computer readable storage medium of example 8, wherein the instructions, when executed, further cause the at least one processor to collect the performance statistic based on the anchor points.
Example 10 includes the at least one non-transitory computer readable storage medium of example 9, wherein the instructions, when executed, further cause the at least one processor to rank features of the previously generated machine learning model to generate the task knowledge.
Example 11 includes the at least one non-transitory computer readable storage medium of example 7, wherein the instructions, when executed, further cause the at least one processor select a prior architecture based on performance of the prior architecture on a selected hardware to create the search space.
Example 12 includes a method for data enhanced automated model generation, the method comprising accessing a request to generate a machine learning model to perform a selected task, generating task knowledge based on a previously generated machine learning model, creating a search space based on the task knowledge, and generating a machine learning model using neural architecture search, the neural architecture search beginning based on the search space.
Example 13 includes the method of example 12, further including, during generation of the machine learning model, inserting a plurality of anchor points into the machine learning model, the anchor points to be used when collecting a performance statistic concerning execution of the machine learning model.
Example 14 includes the method of example 13, further including collecting the performance statistic based on the anchor points.
Example 15 includes the method of example 14, wherein the generation of the task knowledge includes ranking features of the previously generated machine learning model.
Example 16 includes the method of example 12, wherein the creation of the search space includes selecting a prior architecture based on performance of the prior architecture on a selected hardware.
Example 17 includes an apparatus for data enhanced automated model generation, the apparatus comprising means for accessing a request to generate a machine learning model to perform a selected task, means for generating task knowledge based on a previously generated machine learning model, means for creating a search space based on the task knowledge, and means for generating a machine learning model using neural architecture search, the neural architecture search beginning based on the search space.
Example 18 includes the apparatus of example 17, further means for inserting, during generation of the machine learning model, a plurality of anchor points into the machine learning model, the anchor points to be used when collecting a performance statistic concerning execution of the machine learning model.
Example 19 includes the apparatus of example 18, further including means for collecting the performance statistic based on the anchor points.
Example 20 includes the apparatus of example 19, wherein the means for generating is further to rank features of the previously generated machine learning model.
Example 21 includes the apparatus of example 17, wherein the means for creating is to select a prior architecture based on performance of the prior architecture on a selected hardware.
The following claims are hereby incorporated into this Detailed Description by this reference. Although certain example systems, methods, apparatus, and articles of manufacture have been disclosed herein, the scope of coverage of this patent is not limited thereto. On the contrary, this patent covers all systems, methods, apparatus, and articles of manufacture fairly falling within the scope of the claims of this patent.

Claims

What is claimed is:

1. An apparatus for data enhanced automated model generation, the apparatus comprising:

interface circuitry to access a request to generate a machine learning model; and

processor circuitry including one or more of:

at least one of a central processing unit, a graphic processing unit, or a digital signal processor, the at least one of the central processing unit, the graphic processing unit, or the digital signal processor having control circuitry to control data movement within the processor circuitry, arithmetic and logic circuitry to perform one or more first operations corresponding to instructions, and one or more registers to store a result of the one or more first operations, the instructions in the apparatus;

a Field Programmable Gate Array (FPGA), the FPGA including logic gate circuitry, a plurality of configurable interconnections, and storage circuitry, the logic gate circuitry and interconnections to perform one or more second operations, the storage circuitry to store a result of the one or more second operations; or

Application Specific Integrate Circuitry (ASIC) including logic gate circuitry to perform one or more third operations;

the processor circuitry to perform at least one of the first operations, the second operations, or the third operations to instantiate:

task data orchestration circuitry to generate task knowledge based on a previously generated machine learning model;

search space management circuitry to create a search space based on the task knowledge; and

neural architecture search circuitry to generate the machine learning model using neural architecture search, the neural architecture search circuitry to begin an architecture search based on the search space.

2. The apparatus of claim 1, wherein the processor circuitry is to, during generation of the machine learning model, insert a plurality of anchor points into the machine learning model, the anchor points to be used for collection of a performance statistic concerning execution of the machine learning model.

3. The apparatus of claim 2, wherein the performance statistic includes at least one of power efficiency or energy efficiency.

4. The apparatus of claim 2, wherein the processor circuitry is further to collect the performance statistic based on the anchor points.

5. The apparatus of claim 4, wherein, to generate the task knowledge, the processor circuitry is further to rank features of the previously generated machine learning model.

6. The apparatus of claim 1, wherein to create the search space, the processor circuitry is to select a prior architecture based on performance of the prior architecture on a selected hardware.

7. At least one non-transitory computer readable storage medium comprising instructions that, when executed, cause at least one processor to at least:

access a request to generate a machine learning model to perform a selected task;

generate task knowledge based on a previously generated machine learning model;

create a search space based on the task knowledge; and

generate a machine learning model using neural architecture search, the neural architecture search beginning based on the search space.

8. The at least one non-transitory computer readable storage medium of claim 7, wherein the instructions, when executed, further cause the at least one processor to insert a plurality of anchor points into the machine learning model, the anchor points to be used when collecting a performance statistic concerning execution of the machine learning model.

9. The at least one non-transitory computer readable storage medium of claim 8, wherein the instructions, when executed, further cause the at least one processor to collect the performance statistic based on the anchor points.

10. The at least one non-transitory computer readable storage medium of claim 9, wherein the instructions, when executed, further cause the at least one processor to rank features of the previously generated machine learning model to generate the task knowledge.

11. The at least one non-transitory computer readable storage medium of claim 7, wherein the instructions, when executed, further cause the at least one processor select a prior architecture based on performance of the prior architecture on a selected hardware to create the search space.

12. A method for data enhanced automated model generation, the method comprising:

accessing a request to generate a machine learning model to perform a selected task;

generating task knowledge based on a previously generated machine learning model;

creating a search space based on the task knowledge; and

generating a machine learning model using neural architecture search, the neural architecture search beginning based on the search space.

13. The method of claim 12, further including, during generation of the machine learning model, inserting a plurality of anchor points into the machine learning model, the anchor points to be used when collecting a performance statistic concerning execution of the machine learning model.

14. The method of claim 13, further including collecting the performance statistic based on the anchor points.

15. The method of claim 14, wherein the generation of the task knowledge includes ranking features of the previously generated machine learning model.

16. The method of claim 12, wherein the creation of the search space includes selecting a prior architecture based on performance of the prior architecture on a selected hardware.

17. An apparatus for data enhanced automated model generation, the apparatus comprising:

means for accessing a request to generate a machine learning model to perform a selected task;

means for generating task knowledge based on a previously generated machine learning model;

means for creating a search space based on the task knowledge; and

means for generating a machine learning model using neural architecture search, the neural architecture search beginning based on the search space.

18. The apparatus of claim 17, further means for inserting, during generation of the machine learning model, a plurality of anchor points into the machine learning model, the anchor points to be used when collecting a performance statistic concerning execution of the machine learning model.

19. The apparatus of claim 18, further including means for collecting the performance statistic based on the anchor points.

20. The apparatus of claim 19, wherein the means for generating is further to rank features of the previously generated machine learning model.

21. The apparatus of claim 17, wherein the means for creating is to select a prior architecture based on performance of the prior architecture on a selected hardware.