公开(公告)号：US20200373941A1

公开(公告)日：2020-11-26

申请号：US16426303

申请日：2019-05-30

Applicant: Nvidia Corporation

Inventor： Jorge Albericio Latorre ,  Ming Y. Siu

IPC: H03M7/30 ,  G06N3/08

Abstract: In artificial neural networks, and other similar applications, there is typically a large amount of data involved that is considered sparse data. Due to the large size of the data involved in such applications, it is helpful to compress the data to save bandwidth resources when transmitting the data and save memory resources when storing the data. Introduced herein is a compression technique that selects elements with significant values from data and restructures them into a structured sparse format. By generating metadata that enforces the structured sparse format and organizing the data according to the metadata, the introduced technique not only reduces the size of the data but also consistently places the data in a particular format. As such, hardware can be simplified and optimized to process the data much faster and much more efficiently than the conventional compression techniques that rely on a non-structured sparsity format.

公开(公告)号：US11249727B2

公开(公告)日：2022-02-15

申请号：US17073512

申请日：2020-10-19

Applicant: Nvidia Corporation

Inventor： Jorge Albericio Latorre ,  Jeff Pool ,  David Garcia

IPC: G06F7/78 ,  G06F7/57 ,  G06N3/04 ,  G06F16/901 ,  G06F17/16

Abstract: Many computing systems process data organized in a matrix format. For example, artificial neural networks (ANNs) perform numerous computations on data organized into matrices using conventional matrix arithmetic operations. One such operation, which is commonly performed, is the transpose operation. Additionally, many such systems need to process many matrices and/or matrices that are large in size. For sparse matrices that hold few significant values and many values that can be ignored, transmitting and processing all the values in such matrices is wasteful. Thus, techniques are introduced for storing a sparse matrix in a compressed format that allows for a matrix transpose operation to be performed on the compressed matrix without having to first decompress the compressed matrix. By utilizing the introduced techniques, more matrix operations can be performed than conventional systems.

公开(公告)号：US20210034332A1

公开(公告)日：2021-02-04

申请号：US17073512

申请日：2020-10-19

Applicant: Nvidia Corporation

Inventor： Jorge Albericio Latorre ,  Jeff Pool ,  David Garcia

IPC: G06F7/78 ,  G06F7/57 ,  G06N3/04 ,  G06F16/901 ,  G06F17/16

Abstract: Many computing systems process data organized in a matrix format. For example, artificial neural networks (ANNs) perform numerous computations on data organized into matrices using conventional matrix arithmetic operations. One such operation, which is commonly performed, is the transpose operation. Additionally, many such systems need to process many matrices and/or matrices that are large in size. For sparse matrices that hold few significant values and many values that can be ignored, transmitting and processing all the values in such matrices is wasteful. Thus, techniques are introduced for storing a sparse matrix in a compressed format that allows for a matrix transpose operation to be performed on the compressed matrix without having to first decompress the compressed matrix. By utilizing the introduced techniques, more matrix operations can be performed than conventional systems.

公开(公告)号：US20200272425A1

公开(公告)日：2020-08-27

申请号：US16287564

申请日：2019-02-27

Applicant: Nvidia Corporation

Inventor： Jorge Albericio Latorre ,  Jeff Pool ,  David Garcia

IPC: G06F7/78 ,  G06F7/57 ,  G06F17/16 ,  G06F16/901 ,  G06N3/04

Abstract: Many computing systems process data organized in a matrix format. For example, artificial neural networks (ANNs) perform numerous computations on data organized into matrices using conventional matrix arithmetic operations. One such operation, which is commonly performed, is the transpose operation. Additionally, many such systems need to process many matrices and/or matrices that are large in size. For sparse matrices that hold few significant values and many values that can be ignored, transmitting and processing all the values in such matrices is wasteful. Thus, techniques are introduced for storing a sparse matrix in a compressed format that allows for a matrix transpose operation to be performed on the compressed matrix without having to first decompress the compressed matrix. By utilizing the introduced techniques, more matrix operations can be performed than conventional systems.

公开(公告)号：US20250045107A1

公开(公告)日：2025-02-06

申请号：US18240281

申请日：2023-08-30

Applicant: NVIDIA Corporation

Inventor： Jorge Albericio Latorre ,  Chong Yu

IPC: G06F9/50 ,  G06F17/16

Abstract: Apparatuses, systems, and methods to enable matrix multiplication acceleration by modifying an input to apply sparsity through sparse activation filtering. In at least one embodiment, a neural network modifies pixels within an image through sparse activation filtering to enable use of one or more matrix multiplication acceleration units to perform a sparse patch embedding operation.

公开(公告)号：US20230196662A1

公开(公告)日：2023-06-22

申请号：US17556885

申请日：2021-12-20

Applicant: Nvidia Corporation

Inventor： Pietari Kaskela ,  Andrew Tao ,  Michael Ranzinger ,  David Tarjan ,  Jonathan Filip Gustav Granskog ,  Jorge Albericio Latorre

IPC: G06T15/50 ,  G06N3/04 ,  G06T3/40 ,  G06T3/00

CPC classification number: G06T15/503 ,  G06N3/0454 ,  G06T3/4046 ,  G06T3/0093

Abstract: Apparatuses, systems, and techniques are presented to reconstruct one or more images. In at least one embodiment, one or more circuits are to use one or more neural networks to adjust one or more pixel blending weights.

公开(公告)号：US11379420B2

公开(公告)日：2022-07-05

申请号：US16359787

申请日：2019-03-20

Applicant: NVIDIA Corporation

Inventor： Jorge Albericio Latorre ,  Jack H. Choquette ,  Manan Maheshkumar Patel ,  Jeffrey Pool ,  Ming Y. Siu ,  Ronny Meir Krashinsky ,  Ganesh Venkatesh

IPC: G06F16/174 ,  G06F16/901 ,  G06N3/08 ,  H03M7/30 ,  G06F16/14

Abstract: Compressed data is oftentimes beneficial for reducing the computing resources required, for example, to transmit and store data. The compression of data is particularly useful when dealing with sparse data (data that includes numerous zeros or near-zero values) and only non-zero values above a certain threshold have significance. When dealing with compressed data, oftentimes the data needs to be decompressed for processing (e.g., by deep learning networks or other applications configured to operate on sparse, or other uncompressed data). Instructions are disclosed for supporting the decompression of compressed data by a processing unit such as a CPU and GPU.

公开(公告)号：US11489541B2

公开(公告)日：2022-11-01

申请号：US16426303

申请日：2019-05-30

Applicant: Nvidia Corporation

Inventor： Jorge Albericio Latorre ,  Ming Y. Siu

IPC: H03M7/00 ,  H03M7/30 ,  G06N3/08

Abstract: In artificial neural networks, and other similar applications, there is typically a large amount of data involved that is considered sparse data. Due to the large size of the data involved in such applications, it is helpful to compress the data to save bandwidth resources when transmitting the data and save memory resources when storing the data. Introduced herein is a compression technique that selects elements with significant values from data and restructures them into a structured sparse format. By generating metadata that enforces the structured sparse format and organizing the data according to the metadata, the introduced technique not only reduces the size of the data but also consistently places the data in a particular format. As such, hardware can be simplified and optimized to process the data much faster and much more efficiently than the conventional compression techniques that rely on a non-structured sparsity format.

公开(公告)号：US11127167B2

公开(公告)日：2021-09-21

申请号：US16397034

申请日：2019-04-29

Applicant: NVIDIA Corporation

Inventor： Michael Frumkin ,  Jeffrey Pool ,  Jorge Albericio Latorre

IPC: G06K9/36 ,  G06T9/00 ,  G06K9/46

Abstract: Many computing systems process data organized in a matrix format. For example, artificial neural networks perform numerous computations on data organized into matrices using conventional matrix arithmetic operations. One such operation is the transpose operation. Techniques are introduced for storing a matrix in a compressed format that allows, for example, a transpose operation to be performed during decompression. Thus, by utilizing the introduced techniques, transformations of compressed matrices such transposition can be achieved in a more effective way. Parallel processing may also be used to more efficiently compress and/or decompress.

公开(公告)号：US10860293B2

公开(公告)日：2020-12-08

申请号：US16287564

申请日：2019-02-27

Applicant: Nvidia Corporation

Inventor： Jorge Albericio Latorre ,  Jeff Pool ,  David Garcia

IPC: G06F7/78 ,  G06F7/57 ,  G06N3/04 ,  G06F16/901 ,  G06F17/16

Abstract: Many computing systems process data organized in a matrix format. For example, artificial neural networks (ANNs) perform numerous computations on data organized into matrices using conventional matrix arithmetic operations. One such operation, which is commonly performed, is the transpose operation. Additionally, many such systems need to process many matrices and/or matrices that are large in size. For sparse matrices that hold few significant values and many values that can be ignored, transmitting and processing all the values in such matrices is wasteful. Thus, techniques are introduced for storing a sparse matrix in a compressed format that allows for a matrix transpose operation to be performed on the compressed matrix without having to first decompress the compressed matrix. By utilizing the introduced techniques, more matrix operations can be performed than conventional systems.