Patent Grant
11935271
This disclosure relates generally to field of data processing, and more particularly to video encoding and decoding.
ISO/IEC MPEG (JTC 1/SC 29/WG 11) has been actively searching for potential needs for standardization of future video coding technology for visual analysis and understanding. The Compact Descriptors for Visual Search (CDVS) standard was adopted by ISO in 2015 as a still-image standard, which extracts feature representations for image similarity matching. The Compact Descriptors for Visual Analysis (CDVS) standard is listed as Part 15 of MPEG 7 and ISO/IEC 15938-15 and was finalized in 2018, which extracts global and local, hand-designed and DNN-based, feature descriptors for video segments. The success of Deep Neural Network (DNN) in a large range of video applications such as semantic classification, target detection/recognition, target tracking, video quality enhancement, etc. poses a strong need for compressing the DNN models. Therefore, MPEG is actively working on the Coded Representation of Neural Network standard (NNR), which encodes DNN models to save both storage and computation.
There exist several approaches for learning a compact DNN model. The target is to remove unimportant weight coefficients and the assumption is that the smaller the weight coefficients are in value, the less their importance are. Several network pruning methods may be used, which may add sparsity-promoting regularization terms into the network training target. From the perspective of compressing DNN models, after learning a compact network model, the weight coefficients can be further compressed by quantization followed by entropy coding. Such further compression processes can significantly reduce the storage size of the DNN model, which are essential for model deployment over mobile devices, chips, etc.
A selective structured unification method may be used to unify the network weight coefficients for improved quantization and entropy coding in later compression processes. Given a set of pre-defined desired unifying structures, a unification loss value which estimates the error introduced by unifying the weight coefficients according to the unifying structure may be computed. An iterative network retraining/refining process is used to jointly optimize the original training target and the unification loss, so that the learned network weight coefficients preserve the original performance (e.g., optimal cross-entropy loss for classification tasks) and has the optimal desired unifying structure. The design of the desired unifying structure targets at facilitating further compression through quantization and entropy coding. Therefore, the optimized weight coefficients by the method, computer system, and computer readable medium disclosed herein may be more suitable for further compression. The method, computer system, and computer readable medium disclosed herein may be applied to the original pretrained DNN model. It can also be used as an additional processing module to further compress any pruned DNN model.
Embodiments relate to a method, system, and computer readable medium for compressing a neural network model. According to one aspect, a method for compressing a neural network model is provided. The method may include identifying one or more blocks from among a superblock corresponding to a multi-dimensional tensor associated with a neural network. A set of weight coefficients associated with the superblock is unified. A model of the neural network is compressed based on the unified set of weight coefficients.
According to another aspect, a computer system for compressing a neural network model is provided. The computer system may include one or more processors, one or more computer-readable memories, one or more computer-readable tangible storage devices, and program instructions stored on at least one of the one or more storage devices for execution by at least one of the one or more processors via at least one of the one or more memories, whereby the computer system is capable of performing a method. The method may include identifying one or more blocks from among a superblock corresponding to a multi-dimensional tensor associated with a neural network. A set of weight coefficients associated with the superblock is unified. A model of the neural network is compressed based on the unified set of weight coefficients.
According to yet another aspect, a computer readable medium for compressing a neural network model is provided. The computer readable medium may include one or more computer-readable storage devices and program instructions stored on at least one of the one or more tangible storage devices, the program instructions executable by a processor. The program instructions are executable by a processor for performing a method that may accordingly include identifying one or more blocks from among a superblock corresponding to a multi-dimensional tensor associated with a neural network. A set of weight coefficients associated with the superblock is unified. A model of the neural network is compressed based on the unified set of weight coefficients.
These and other objects, features and advantages will become apparent from the following detailed description of illustrative embodiments, which is to be read in connection with the accompanying drawings. The various features of the drawings are not to scale as the illustrations are for clarity in facilitating the understanding of one skilled in the art in conjunction with the detailed description. In the drawings:
Detailed embodiments of the claimed structures and methods are disclosed herein; however, it can be understood that the disclosed embodiments are merely illustrative of the claimed structures and methods that may be embodied in various forms. Those structures and methods may, however, be embodied in many different forms and should not be construed as limited to the exemplary embodiments set forth herein. Rather, these exemplary embodiments are provided so that this disclosure will be thorough and complete and will fully convey the scope to those skilled in the art. In the description, details of well-known features and techniques may be omitted to avoid unnecessarily obscuring the presented embodiments.
Embodiments relate generally to the field of data processing, and more particularly to neural networks. The following described exemplary embodiments provide a system, method and computer program to, among other things, compress neural network models based on selective structured weight unification in an iterative network. The structured weight unification improves the efficiency of further compression of the learned weight coefficients, which can significantly reduce the DNN model size. The unifying structure selection using the unification loss may enable the system to use flexible unification structure to reduce the unification loss with similar compression efficiency. Through the iterative retraining process, the original training target can be maintained for improved compression efficiency. The iterative retraining process also gives the flexibility of introducing different loss at different times, making the system focus on different target during the optimization process. Both dense and pruned sparse models may be used, which serves as an additional module to improve model efficiency in later compression and computation. The method, computer system, and computer readable medium may be applied to datasets with different data forms. The input/output data are general 4D tensors, which can be real video segments, images, or extracted feature maps
Aspects are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer readable media according to the various embodiments. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer readable program instructions.
Referring now to
The system 100 may include a computer 102 and a server computer 114. The computer 102 may communicate with the server computer 114 via a communication network 110 (hereinafter “network”). The computer 102 may include a processor 104 and a software program 108 that is stored on a data storage device 106 and is enabled to interface with a user and communicate with the server computer 114. As will be discussed below with reference to
The server computer 114 may also operate in a cloud computing service model, such as Software as a Service (SaaS), Platform as a Service (PaaS), or Infrastructure as a Service (IaaS), as discussed below with respect to
The server computer 114, which may be used for compressing a neural network model is enabled to run a Neural Network Model Compression Program 116 (hereinafter “program”) that may interact with a database 112. The Neural Network Model Compression Program method is explained in more detail below with respect to
It should be noted, however, that processing for the program 116 may, in some instances be shared amongst the computers 102 and the server computers 114 in any ratio. In another embodiment, the program 116 may operate on more than one computer, server computer, or some combination of computers and server computers, for example, a plurality of computers 102 communicating across the network 110 with a single server computer 114. In another embodiment, for example, the program 116 may operate on a plurality of server computers 114 communicating across the network 110 with a plurality of client computers. Alternatively, the program may operate on a network server communicating across the network with a server and a plurality of client computers.
The network 110 may include wired connections, wireless connections, fiber optic connections, or some combination thereof. In general, the network 110 can be any combination of connections and protocols that will support communications between the computer 102 and the server computer 114. The network 110 may include various types of networks, such as, for example, a local area network (LAN), a wide area network (WAN) such as the Internet, a telecommunication network such as the Public Switched Telephone Network (PSTN), a wireless network, a public switched network, a satellite network, a cellular network (e.g., a fifth generation (5G) network, a long-term evolution (LTE) network, a third generation (3G) network, a code division multiple access (CDMA) network, etc.), a public land mobile network (PLMN), a metropolitan area network (MAN), a private network, an ad hoc network, an intranet, a fiber optic-based network, or the like, and/or a combination of these or other types of networks.
The number and arrangement of devices and networks shown in
Referring now to
={(x,y)} may denote a data set where a target y is assigned to an input x. ΘO={WO} may denote a set of weight coefficients of a DNN, where WO denotes the weight coefficients of a network layer. The target of the neural network training is to learn an optimal set of weight coefficients Θ={W} so that a loss £(
|Θ) can be minimized. For example, in previous network pruning approaches, the loss £(
|Θ) has two parts, an empirical data loss £
(
|Θ) such as the cross-entropy loss for classification tasks, and a sparsity-promoting regularization loss £R(O):
£(|Θ)=£D(
|Θ)+λR£R(Θ), (Eq. 1)
where λR is a hyperparameter balancing the contributions of the data loss and the regularization loss.
The sparsity-promoting regularization loss places regularization over the entire weight coefficients, and the resulting sparse weights have weak relationship with the inference efficiency or computation acceleration. From another perspective, after pruning, the sparse weights can further go through another network training process where an updated optimal set of weight coefficients can be learned that can improve the quantization efficiency for further model compression.
A unification process may occur over the weight coefficients Θ={W}, where W may denote the weight coefficients of a network layer. W is a general 5-Dimension (5D) tensor with size (ci, k1, k2, k3, co). The input of the layer is a 4-Dimension (4D) tensor A of size (hi,wi,di,ci), and the output of the layer is a 4D tensor B of size (ho,wo,do,co). The sizes ci, k1, k2, k3, co, hi, wi, di, ho, wo, do are integer numbers greater or equal to 1. When any of the sizes ci, k1, k2, k3, co, hi, wi, di, ho, wo, do takes number 1, the corresponding tensor reduces to a lower dimension. Each item in each tensor is a floating number. M may denote a 5D binary mask of the same size as W,
where each item in M is a binary number 0/1 indicating whether the corresponding weight coefficient is pruned/kept. M is introduced to be associated with W to cope with the case where W is from a pruned DNN model. When W is from a dense unpruned DNN model, all items in M take value 1. The output B is computed through the convolution operation ⊙ based on A, M and W:
The parameters hi, wi and di (h0, wo and do) are the height, weight and depth of the input tensor A (output tensor B). The parameter ci (co) is the number of input (output) channel. The parameters k1, k2 and k3 are the size of the convolution kernel corresponding to the height, weight and depth axes, respectively. That is, for each output channel v=1, . . . , co, the operation can be seen as a 4D weight tensor Wv of size (ci,k1,k2,k3) convolving with the input A.
The order of the summation operation can be changed, and such a change of the order of the summation operation corresponds to reshaping the weight tensor (and the corresponding input and output tensors). In one or more embodiments, the 5D weight tensor may be reshaped into a 3D tensor of size (ci,co,k) where k=k1·k2·k3.
The reshaped 3D weight tensor Wreshape can be further partitioned into different structures. For example, Wreshape can be partitioned into 2D or 3D super-blocks, and each super-block can be divided into 2D or 3D blocks. The unification process may occur within each super-block. That is, a unification loss is computed over each super-block, and an optimal unifying structure is determined for each super-block based on the unification loss. Specifically, a set of desired unifying structure is pre-defined, by taking into consideration of two aspects. The unifying structure of the weight coefficients should align with the underlying GEMM matrix multiplication process of how the convolution operation is implemented so that the inference computation of using the unified weight coefficients is accelerated. The unifying structure of the weight coefficients can help improve the quantization and entropy coding efficiency.
The 3D weight tensor Wreshape may be partitioned into super-blocks of size (gi,go,gk), and each super-block may be further partitioned into blocks. According to one or more embodiments, gi=4, go=4, gk=1, and the super-block reduces to a 2D matrix. Each block may have 4 weight coefficients, and the weights within a block may be unified.
S may denote a super-block, and f(S) may denote a way of partitioning the super-block S. A loss £U(f(S)) can be computed to measure the error introduced by unifying weight coefficients according to f(S):
£U(f(S))=ΣB∈f(S)LU(B), (Eq. 3)
where LU(B) measures the loss of unifying weight coefficients in a block denoted by B. To unify a block, the absolute value of all weight coefficients in B may be set to be the same, while keeping the original sign of the weight coefficients. That is, a weight coefficient wi is set to be +/−VB and the sign +/− is the same with the original wi. In one or more embodiments, VB=mean(abs(B)), (i.e., the mean of the absolute values of weights in B) and LU(B)=std(abs(B)) (i.e., the standard deviation of the absolute values of weights in B). For each super-block S, the optimal way of partition f*(S) can be selected as the one with the minimal loss:
f*(S)=argminf(S)£U(f(S)). (Eq. 4)
Referring now to
An iterative selective structured weight optimization process may be used to learn the updated optimal weight coefficients Θ*={W*} through a neural network retraining/refining process. In the framework of the iterative optimization process, both the target training loss and the structured unification loss may be iteratively optimized. Specifically, the input to the system may be a pre-trained DNN model with weight coefficients Θ={W} and corresponding masks {M}, which can be either a pruned sparse model or an un-pruned dense model. W and M denote the weight coefficients and corresponding mask for one layer in the pre-trained DNN model. The optimal unifying structure {f*(S)} of W and M may be determined for the set of super-blocks {S} in W, using the unifying structure selection module 302. A list of super-blocks may be determined and the weight coefficients of these super-blocks may be unified by the weight unification module 304. To determine the list of super-blocks, all the loss values LU(f*(S)) may be collected for all super-blocks in W, and the super-blocks may be ranked based on LU(f*(S)) in ascending order. Given a unifying ratio q as a hyperparameter, the first q percent super-blocks of W are selected, and the corresponding weights are unified. The output of the weight unification module 304 may be a unified weight WU and a corresponding unifying mask MU. MU has the same shape with WU, and each entry in MU indicates whether the corresponding weight in WU is unified.
Given the training data set {(x,y)}, where
can be the same as the original dataset
0={(x0,y0)} based on which the pre-trained weight coefficients W are obtained.
can also be a different dataset from
0, but with the same data distribution as the original dataset
. The unified weight coefficients of WU marked in MU are fixed, and a neural network training process may be performed to update the remaining un-fixed weight coefficients in WU, e.g., through the back-propagation mechanism. The hyperparameter q can be pre-determined or can change (e.g., increasing) so that more weight coefficients can be fixed and unified gradually for later iterations during training, to avoid premature unification. For the t-th iteration, the current unified weight coefficients WU(t) and unifying mask MU(t) may be used. The target of this step is to find a set of updated optimal weight coefficients WU* and pruning mask M* by iteratively minimizing the total loss. Specifically, each input x is passed though the network via the network forward computation module 308 using the current unified weight coefficients WU(t), which generates an estimated output
|Θ) can be computed through the compute training loss module 310. The gradient of the loss G(WU(t)) can be computed by the compute gradient module 312. The automatic gradient computing method used by deep learning frameworks such as tensorflow or pytorch can be used here to compute G(WU(t)). Based on the gradient G(WU(t)) and the unifying mask MU(t), the non-fixed weight coefficients of WU(t) and the corresponding items in the pruning mask M(t) can be updated through back-propagation using the back propagation and weight update module 314. The retraining module 306 may include an iterative process, in which multiple iterations are taken to update the non-fixed parts of WU(t) and the corresponding pruning mask M(t), e.g., until the training converges. The system may move to the next iteration t+1, where given a new unifying ratio q(t+1), based on WU(t), a new unified weight coefficients WU(t+1) and the corresponding unifying mask MU(t+1) can be computed, and the system goes to the next iteration.
Referring now to
At 402, the method 400 includes identifying one or more blocks from among a superblock corresponding to a multi-dimensional tensor associated with a neural network.
At 404, the method 400 includes unifying a set of weight coefficients associated with the superblock.
At 406, the method 400 includes compressing a model of the neural network based on the unified set of weight coefficients.
It may be appreciated that
Computer 102 (
Processor 820 is implemented in hardware, firmware, or a combination of hardware and software. Processor 820 is a central processing unit (CPU), a graphics processing unit (GPU), an accelerated processing unit (APU), a microprocessor, a microcontroller, a digital signal processor (DSP), a field-programmable gate array (FPGA), an application-specific integrated circuit (ASIC), or another type of processing component. In some implementations, processor 820 includes one or more processors capable of being programmed to perform a function. Bus 826 includes a component that permits communication among the internal components 800A,B.
The one or more operating systems 828, the software program 108 (
Each set of internal components 800A,B also includes a R/W drive or interface 832 to read from and write to one or more portable computer-readable tangible storage devices 936 such as a CD-ROM, DVD, memory stick, magnetic tape, magnetic disk, optical disk or semiconductor storage device. A software program, such as the software program 108 (
Each set of internal components 800A,B also includes network adapters or interfaces 836 such as a TCP/IP adapter cards; wireless Wi-Fi interface cards; or 3G, 4G, or 5G wireless interface cards or other wired or wireless communication links. The software program 108 (
Each of the sets of external components 900A,B can include a computer display monitor 920, a keyboard 930, and a computer mouse 934. External components 900A,B can also include touch screens, virtual keyboards, touch pads, pointing devices, and other human interface devices. Each of the sets of internal components 800A,B also includes device drivers 840 to interface to computer display monitor 920, keyboard 930 and computer mouse 934. The device drivers 840, R/W drive or interface 832 and network adapter or interface 836 comprise hardware and software (stored in storage device 830 and/or ROM 824).
It is understood in advance that although this disclosure includes a detailed description on cloud computing, implementation of the teachings recited herein are not limited to a cloud computing environment. Rather, some embodiments are capable of being implemented in conjunction with any other type of computing environment now known or later developed.
Cloud computing is a model of service delivery for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g. networks, network bandwidth, servers, processing, memory, storage, applications, virtual machines, and services) that can be rapidly provisioned and released with minimal management effort or interaction with a provider of the service. This cloud model may include at least five characteristics, at least three service models, and at least four deployment models.
Characteristics are as follows:
On-demand self-service: a cloud consumer can unilaterally provision computing capabilities, such as server time and network storage, as needed automatically without requiring human interaction with the service's provider.
Broad network access: capabilities are available over a network and accessed through standard mechanisms that promote use by heterogeneous thin or thick client platforms (e.g., mobile phones, laptops, and PDAs).
Resource pooling: the provider's computing resources are pooled to serve multiple consumers using a multi-tenant model, with different physical and virtual resources dynamically assigned and reassigned according to demand. There is a sense of location independence in that the consumer generally has no control or knowledge over the exact location of the provided resources but may be able to specify location at a higher level of abstraction (e.g., country, state, or datacenter).
Rapid elasticity: capabilities can be rapidly and elastically provisioned, in some cases automatically, to quickly scale out and rapidly released to quickly scale in. To the consumer, the capabilities available for provisioning often appear to be unlimited and can be purchased in any quantity at any time.
Measured service: cloud systems automatically control and optimize resource use by leveraging a metering capability at some level of abstraction appropriate to the type of service (e.g., storage, processing, bandwidth, and active user accounts). Resource usage can be monitored, controlled, and reported providing transparency for both the provider and consumer of the utilized service.
Service Models are as follows:
Software as a Service (SaaS): the capability provided to the consumer is to use the provider's applications running on a cloud infrastructure. The applications are accessible from various client devices through a thin client interface such as a web browser (e.g., web-based e-mail). The consumer does not manage or control the underlying cloud infrastructure including network, servers, operating systems, storage, or even individual application capabilities, with the possible exception of limited user-specific application configuration settings.
Platform as a Service (PaaS): the capability provided to the consumer is to deploy onto the cloud infrastructure consumer-created or acquired applications created using programming languages and tools supported by the provider. The consumer does not manage or control the underlying cloud infrastructure including networks, servers, operating systems, or storage, but has control over the deployed applications and possibly application hosting environment configurations.
Infrastructure as a Service (IaaS): the capability provided to the consumer is to provision processing, storage, networks, and other fundamental computing resources where the consumer is able to deploy and run arbitrary software, which can include operating systems and applications. The consumer does not manage or control the underlying cloud infrastructure but has control over operating systems, storage, deployed applications, and possibly limited control of select networking components (e.g., host firewalls).
Deployment Models are as follows:
Private cloud: the cloud infrastructure is operated solely for an organization. It may be managed by the organization or a third party and may exist on-premises or off-premises.
Community cloud: the cloud infrastructure is shared by several organizations and supports a specific community that has shared concerns (e.g., mission, security requirements, policy, and compliance considerations). It may be managed by the organizations or a third party and may exist on-premises or off-premises.
Public cloud: the cloud infrastructure is made available to the general public or a large industry group and is owned by an organization selling cloud services.
Hybrid cloud: the cloud infrastructure is a composition of two or more clouds (private, community, or public) that remain unique entities but are bound together by standardized or proprietary technology that enables data and application portability (e.g., cloud bursting for load-balancing between clouds).
A cloud computing environment is service oriented with a focus on statelessness, low coupling, modularity, and semantic interoperability. At the heart of cloud computing is an infrastructure comprising a network of interconnected nodes.
Referring to
Referring to
Hardware and software layer 60 includes hardware and software components. Examples of hardware components include: mainframes 61; RISC (Reduced Instruction Set Computer) architecture based servers 62; servers 63; blade servers 64; storage devices 65; and networks and networking components 66. In some embodiments, software components include network application server software 67 and database software 68.
Virtualization layer 70 provides an abstraction layer from which the following examples of virtual entities may be provided: virtual servers 71; virtual storage 72; virtual networks 73, including virtual private networks; virtual applications and operating systems 74; and virtual clients 75.
In one example, management layer 80 may provide the functions described below. Resource provisioning 81 provides dynamic procurement of computing resources and other resources that are utilized to perform tasks within the cloud computing environment. Metering and Pricing 82 provide cost tracking as resources are utilized within the cloud computing environment, and billing or invoicing for consumption of these resources. In one example, these resources may comprise application software licenses. Security provides identity verification for cloud consumers and tasks, as well as protection for data and other resources. User portal 83 provides access to the cloud computing environment for consumers and system administrators. Service level management 84 provides cloud computing resource allocation and management such that required service levels are met. Service Level Agreement (SLA) planning and fulfillment 85 provide pre-arrangement for, and procurement of, cloud computing resources for which a future requirement is anticipated in accordance with an SLA.
Workloads layer 90 provides examples of functionality for which the cloud computing environment may be utilized. Examples of workloads and functions which may be provided from this layer include: mapping and navigation 91; software development and lifecycle management 92; virtual classroom education delivery 93; data analytics processing 94; transaction processing 95; and Neural Network Model Compression 96. Neural Network Model Compression 96 may compress a neural network model based on selective structured weight unification in an iterative network.
Some embodiments may relate to a system, a method, and/or a computer readable medium at any possible technical detail level of integration. The computer readable medium may include a computer-readable non-transitory storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out operations.
The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.
Computer readable program instructions described herein can be downloaded to respective computing/processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and/or a wireless network. The network may comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. A network adapter card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing/processing device.
Computer readable program code/instructions for carrying out operations may be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, configuration data for integrated circuitry, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++, or the like, and procedural programming languages, such as the “C” programming language or similar programming languages. The computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), or programmable logic arrays (PLA) may execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects or operations.
These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including instructions which implement aspects of the function/act specified in the flowchart and/or block diagram block or blocks.
The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process, such that the instructions which execute on the computer, other programmable apparatus, or other device implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer readable media according to various embodiments. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). The method, computer system, and computer readable medium may include additional blocks, fewer blocks, different blocks, or differently arranged blocks than those depicted in the Figures. In some alternative implementations, the functions noted in the blocks may occur out of the order noted in the Figures. For example, two blocks shown in succession may, in fact, be executed concurrently or substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts or carry out combinations of special purpose hardware and computer instructions.
It will be apparent that systems and/or methods, described herein, may be implemented in different forms of hardware, firmware, or a combination of hardware and software. The actual specialized control hardware or software code used to implement these systems and/or methods is not limiting of the implementations. Thus, the operation and behavior of the systems and/or methods were described herein without reference to specific software code—it being understood that software and hardware may be designed to implement the systems and/or methods based on the description herein.
No element, act, or instruction used herein should be construed as critical or essential unless explicitly described as such. Also, as used herein, the articles “a” and “an” are intended to include one or more items, and may be used interchangeably with “one or more.” Furthermore, as used herein, the term “set” is intended to include one or more items (e.g., related items, unrelated items, a combination of related and unrelated items, etc.), and may be used interchangeably with “one or more.” Where only one item is intended, the term “one” or similar language is used. Also, as used herein, the terms “has,” “have,” “having,” or the like are intended to be open-ended terms. Further, the phrase “based on” is intended to mean “based, at least in part, on” unless explicitly stated otherwise.
The descriptions of the various aspects and embodiments have been presented for purposes of illustration, but are not intended to be exhaustive or limited to the embodiments disclosed. Even though combinations of features are recited in the claims and/or disclosed in the specification, these combinations are not intended to limit the disclosure of possible implementations. In fact, many of these features may be combined in ways not specifically recited in the claims and/or disclosed in the specification. Although each dependent claim listed below may directly depend on only one claim, the disclosure of possible implementations includes each dependent claim in combination with every other claim in the claim set. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope of the described embodiments. The terminology used herein was chosen to best explain the principles of the embodiments, the practical application or technical improvement over technologies found in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
This application claims priority from U.S. Provisional Patent Application No. 62/959,631, filed on Jan. 10, 2020 in the U.S. Patent and Trademark Office, which is incorporated herein by reference in its entirety.
| Number | Name | Date | Kind |
|---|---|---|---|
| 11468313 | Naumov | Oct 2022 | B1 |
| 20180046895 | Xie | Feb 2018 | A1 |
| 20180046914 | Li | Feb 2018 | A1 |
| 20180046915 | Sun | Feb 2018 | A1 |
| 20180046919 | Li | Feb 2018 | A1 |
| 20180082181 | Brothers | Mar 2018 | A1 |
| 20180197081 | Ji | Jul 2018 | A1 |
| 20180253401 | Cardinaux | Sep 2018 | A1 |
| 20180285736 | Baum | Oct 2018 | A1 |
| 20190080238 | Wang | Mar 2019 | A1 |
| 20190122116 | Choi | Apr 2019 | A1 |
| 20190138882 | Choi | May 2019 | A1 |
| 20190251445 | Movshovitz-Attias et al. | Aug 2019 | A1 |
| 20190362235 | Xu | Nov 2019 | A1 |
| 20200082269 | Gao | Mar 2020 | A1 |
| 20210073643 | Partovi Nia | Mar 2021 | A1 |
| 20210081798 | Cho | Mar 2021 | A1 |
| 20210125070 | Wang | Apr 2021 | A1 |
| 20210217204 | Jiang | Jul 2021 | A1 |
| 20210266607 | Jiang | Aug 2021 | A1 |
| 20220067527 | Xu | Mar 2022 | A1 |
| 20220129759 | Yao | Apr 2022 | A1 |
| Number | Date | Country |
|---|---|---|
| 109948794 | Jun 2019 | CN |
| Entry |
|---|
| Ye, S., Zhang, T., Zhang, K., Li, J., Xie, J., Liang, Y., Liu, S., Lin, X., and Wang, Y. A unified framework of dnn weight pruning and weight clustering/quantization using admm. arXiv preprint arXiv:1811.01907 (2018). (Year: 2018). |
| A. Ren, J. Li, T. Zhang, S. Ye, W. Xu, X. Qian, X. Lin, and Y. Wang. Admm-nn: An algorithm-hardware co-design framework of dnns using alternating direction methods of multiplierss. International conference on Architectural Support for Programming Languages and Operating Systems, 2019. (Year: 2019). |
| https://realpython.com/gradient-descent-algorithm-python/ (Year: 2023). |
| Song Han et al., “Learning both Weights and Connections for Efficient Neural Networks”, 9 pages. |
| Number | Date | Country | |
|---|---|---|---|
| 20210217204 A1 | Jul 2021 | US |
| Number | Date | Country | |
|---|---|---|---|
| 62959631 | Jan 2020 | US |
