GB2556413B - Exploiting input data sparsity in neural network compute units - Google Patents

Exploiting input data sparsity in neural network compute units Download PDF

Info

Publication number
GB2556413B
GB2556413B GB201715032A GB201715032A GB2556413B GB 2556413 B GB2556413 B GB 2556413B GB 201715032 A GB201715032 A GB 201715032A GB 201715032 A GB201715032 A GB 201715032A GB 2556413 B GB2556413 B GB 2556413B
Authority
GB
United Kingdom
Prior art keywords
neural network
input data
compute units
network compute
data sparsity
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
GB201715032A
Other versions
GB2556413A (en
GB201715032D0 (en
Inventor
Hyuk Woo Dong
Narayanaswami Ravi
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Google LLC
Original Assignee
Google LLC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from US15/336,066 external-priority patent/US10360163B2/en
Application filed by Google LLC filed Critical Google LLC
Publication of GB201715032D0 publication Critical patent/GB201715032D0/en
Publication of GB2556413A publication Critical patent/GB2556413A/en
Application granted granted Critical
Publication of GB2556413B publication Critical patent/GB2556413B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/15—Correlation function computation including computation of convolution operations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
    • G06N3/063—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F15/00—Digital computers in general; Data processing equipment in general
    • G06F15/76—Architectures of general purpose stored program computers
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30003—Arrangements for executing specific machine instructions
    • G06F9/30007—Arrangements for executing specific machine instructions to perform operations on data operands
    • G06F9/3001—Arithmetic instructions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/38—Concurrent instruction execution, e.g. pipeline or look ahead
    • G06F9/3824—Operand accessing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Computing Systems (AREA)
  • Biophysics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Neurology (AREA)
  • Computer Hardware Design (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Advance Control (AREA)
GB201715032A 2016-10-27 2017-09-19 Exploiting input data sparsity in neural network compute units Active GB2556413B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US15/336,066 US10360163B2 (en) 2016-10-27 2016-10-27 Exploiting input data sparsity in neural network compute units
US15/465,774 US9818059B1 (en) 2016-10-27 2017-03-22 Exploiting input data sparsity in neural network compute units

Publications (3)

Publication Number Publication Date
GB201715032D0 GB201715032D0 (en) 2017-11-01
GB2556413A GB2556413A (en) 2018-05-30
GB2556413B true GB2556413B (en) 2020-01-01

Family

ID=59772734

Family Applications (1)

Application Number Title Priority Date Filing Date
GB201715032A Active GB2556413B (en) 2016-10-27 2017-09-19 Exploiting input data sparsity in neural network compute units

Country Status (1)

Country Link
GB (1) GB2556413B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111028512B (en) * 2019-12-31 2021-06-04 福建工程学院 Real-time traffic prediction method and device based on sparse BP neural network

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5267185A (en) * 1989-04-14 1993-11-30 Sharp Kabushiki Kaisha Apparatus for calculating matrices
US20120143932A1 (en) * 2010-12-06 2012-06-07 International Business Machines Corporation Data Structure For Tiling And Packetizing A Sparse Matrix

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5267185A (en) * 1989-04-14 1993-11-30 Sharp Kabushiki Kaisha Apparatus for calculating matrices
US20120143932A1 (en) * 2010-12-06 2012-06-07 International Business Machines Corporation Data Structure For Tiling And Packetizing A Sparse Matrix

Also Published As

Publication number Publication date
GB2556413A (en) 2018-05-30
GB201715032D0 (en) 2017-11-01

Similar Documents

Publication Publication Date Title
HK1254700A1 (en) Exploiting input data sparsity in neural network compute units
GB201803806D0 (en) Transposing neural network Matrices in hardware
IL261245A (en) Structure learning in convolutional neural networks
GB2538022B (en) Multiple secrets in quorum based data processing
GB201802414D0 (en) Implementing neural networks in fixed point arithmetric computing systems
GB201917993D0 (en) Neural network classification
GB2552577B (en) Neuromorphic synapses
SG11201610016QA (en) Data distribution fabric in scalable gpus
EP3233313C0 (en) Sorting system with multiple sorting devices and corresponding method
GB2568102B (en) Exploiting sparsity in a neural network
ZA201902904B (en) Enabling multiple numerologies in a network
PL3409083T3 (en) Improvements in and relating to data centres
GB201717309D0 (en) Generating randomness in neural networks
GB2561396B (en) Data security using two operating environments operating in parallel
EP3149695A4 (en) Predicting social, economic, and learning outcomes
GB201620232D0 (en) Data input system with online learning
GB201700734D0 (en) Modular computer system and server module
IL261888A (en) Rule enforcement in a network
GB201712817D0 (en) Repartitioning data in distributed computing system
GB2592165B (en) Efficient resource utilization in data centers
GB2545299B (en) Efficient support for variable width data channels in an interconnect network
GB201802417D0 (en) Optical computing devices and methods utilzing multiple integrated computatational elements in sequence
GB201721574D0 (en) Improvements in hair-cutting devices
ZA201702039B (en) Trust management in transaction systems
EP3403212C0 (en) Password maintenance in computer networks