BR112017026917A2 - ?método e dispositivo para processamento de rede neural convolucional cnn e mídia de armazenamento não volátil? - Google Patents

?método e dispositivo para processamento de rede neural convolucional cnn e mídia de armazenamento não volátil?

Info

Publication number
BR112017026917A2
BR112017026917A2 BR112017026917-1A BR112017026917A BR112017026917A2 BR 112017026917 A2 BR112017026917 A2 BR 112017026917A2 BR 112017026917 A BR112017026917 A BR 112017026917A BR 112017026917 A2 BR112017026917 A2 BR 112017026917A2
Authority
BR
Brazil
Prior art keywords
module
asic
cnn
intensity
arithmetic device
Prior art date
Application number
BR112017026917-1A
Other languages
English (en)
Chinese (zh)
Inventor
Zhu Ben
Original Assignee
Tencent Technology (Shenzhen) Company Limited
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
Application filed by Tencent Technology (Shenzhen) Company Limited filed Critical Tencent Technology (Shenzhen) Company Limited
Publication of BR112017026917A2 publication Critical patent/BR112017026917A2/pt

Links

Classifications

    • 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
    • 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
    • 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
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G06F9/5044—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering hardware capabilities
    • 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
    • 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
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • 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
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/50—Indexing scheme relating to G06F9/50
    • G06F2209/501—Performance criteria
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

trata-se de um método e um dispositivo para processamento de cnn para uso na realização de computação de módulo por um modelo de cnn em um asic e para garantir desempenho computacional de cnn e custos de computação menores. o método compreende: adquirir um tipo de intensidade de um primeiro módulo em um modelo de cnn (101); caso o tipo de intensidade do primeiro módulo seja uma intensidade computacional, instalar o primeiro módulo em um asic e adquirir uma pluralidade de recursos de dispositivo aritméticos do asic ocupados pelo primeiro módulo (102); combinar os mesmos recursos de dispositivo aritméticos da pluralidade de recursos de dispositivo aritméticos do asic ocupados pelo primeiro módulo de modo a adquirir um primeiro módulo de recurso combinado (103); e operar o primeiro módulo de recurso combinado no asic (104).
BR112017026917-1A 2016-01-12 2017-01-09 ?método e dispositivo para processamento de rede neural convolucional cnn e mídia de armazenamento não volátil? BR112017026917A2 (pt)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
CN201610017755.1 2016-01-12
CN201610017755.1A CN105678379B (zh) 2016-01-12 2016-01-12 一种cnn的处理方法和装置
PCT/CN2017/070628 WO2017121297A1 (zh) 2016-01-12 2017-01-09 Cnn的处理方法和装置

Publications (1)

Publication Number Publication Date
BR112017026917A2 true BR112017026917A2 (pt) 2018-08-14

Family

ID=56300134

Family Applications (1)

Application Number Title Priority Date Filing Date
BR112017026917-1A BR112017026917A2 (pt) 2016-01-12 2017-01-09 ?método e dispositivo para processamento de rede neural convolucional cnn e mídia de armazenamento não volátil?

Country Status (10)

Country Link
US (1) US11151442B2 (pt)
EP (1) EP3404587B1 (pt)
JP (1) JP6507271B2 (pt)
KR (1) KR102192468B1 (pt)
CN (1) CN105678379B (pt)
BR (1) BR112017026917A2 (pt)
CA (1) CA2987325C (pt)
MY (1) MY188759A (pt)
SG (1) SG11201709689PA (pt)
WO (1) WO2017121297A1 (pt)

Families Citing this family (33)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105678379B (zh) 2016-01-12 2020-08-07 腾讯科技（深圳）有限公司 一种cnn的处理方法和装置
CN106228240B (zh) * 2016-07-30 2020-09-01 复旦大学 基于fpga的深度卷积神经网络实现方法
KR20180034853A (ko) * 2016-09-28 2018-04-05 에스케이하이닉스 주식회사 합성곱 신경망의 연산 장치 및 방법
US10733505B2 (en) * 2016-11-10 2020-08-04 Google Llc Performing kernel striding in hardware
CN108205700B (zh) * 2016-12-20 2021-07-30 上海寒武纪信息科技有限公司 神经网络运算装置和方法
US11238334B2 (en) 2017-04-04 2022-02-01 Hailo Technologies Ltd. System and method of input alignment for efficient vector operations in an artificial neural network
US10387298B2 (en) 2017-04-04 2019-08-20 Hailo Technologies Ltd Artificial neural network incorporating emphasis and focus techniques
US11544545B2 (en) 2017-04-04 2023-01-03 Hailo Technologies Ltd. Structured activation based sparsity in an artificial neural network
US11551028B2 (en) 2017-04-04 2023-01-10 Hailo Technologies Ltd. Structured weight based sparsity in an artificial neural network
US11615297B2 (en) 2017-04-04 2023-03-28 Hailo Technologies Ltd. Structured weight based sparsity in an artificial neural network compiler
US10467795B2 (en) * 2017-04-08 2019-11-05 Intel Corporation Sub-graph in frequency domain and dynamic selection of convolution implementation on a GPU
CN107491809B (zh) * 2017-08-21 2020-10-16 苏州浪潮智能科技有限公司 一种fpga实现残差网络中激活函数的方法
CN108021938A (zh) * 2017-11-29 2018-05-11 中冶南方工程技术有限公司 一种冷轧带钢表面缺陷在线检测方法以及检测系统
JP6574004B2 (ja) * 2018-01-25 2019-09-11 株式会社モルフォ 計算手法決定システム、計算手法決定装置、処理装置、計算手法決定方法、処理方法、計算手法決定プログラム、及び、処理プログラム
CN108470211B (zh) * 2018-04-09 2022-07-12 郑州云海信息技术有限公司 一种卷积计算的实现方法、设备和计算机存储介质
CN110533666B (zh) * 2018-05-25 2022-09-23 杭州海康威视数字技术股份有限公司 一种获取数据块尺寸的方法、处理数据的方法及装置
CN108805285B (zh) * 2018-05-30 2022-03-29 山东浪潮科学研究院有限公司 一种卷积神经网络池化单元设计方法
AU2019295631B2 (en) * 2018-06-27 2022-03-24 Amazon Technologies, Inc. Attached accelerator based inference service
US11960935B2 (en) 2018-06-27 2024-04-16 Amazon Technologies, Inc. Fault-tolerant accelerator based inference service
CN109165736B (zh) * 2018-08-08 2023-12-12 北京字节跳动网络技术有限公司 应用于卷积神经网络的信息处理方法和装置
KR101984730B1 (ko) * 2018-10-23 2019-06-03 (주) 글루시스 서버 장애 자동 예측 시스템 및 자동 예측 방법
CN111488216B (zh) * 2019-01-28 2024-04-30 北京灵汐科技有限公司 一种数据处理的方法、装置及电子设备
CN110032374B (zh) * 2019-03-21 2023-04-07 深兰科技（上海）有限公司 一种参数提取方法、装置、设备及介质
CN110555519B (zh) * 2019-10-08 2023-07-28 东南大学 一种基于符号随机计算的低复杂度卷积神经网络架构
CN111767059B (zh) * 2020-06-30 2023-03-31 北京百度网讯科技有限公司 深度学习模型的部署方法、装置、电子设备和存储介质
CN111915016B (zh) * 2020-07-10 2022-03-25 深圳云天励飞技术股份有限公司 一种基于tvm编译器的异构平台的部署方法及装置
CN111984864B (zh) * 2020-08-14 2023-12-26 北京达佳互联信息技术有限公司 对象推荐方法、装置、电子设备和存储介质
US11221929B1 (en) 2020-09-29 2022-01-11 Hailo Technologies Ltd. Data stream fault detection mechanism in an artificial neural network processor
US11811421B2 (en) 2020-09-29 2023-11-07 Hailo Technologies Ltd. Weights safety mechanism in an artificial neural network processor
US11237894B1 (en) 2020-09-29 2022-02-01 Hailo Technologies Ltd. Layer control unit instruction addressing safety mechanism in an artificial neural network processor
US11263077B1 (en) 2020-09-29 2022-03-01 Hailo Technologies Ltd. Neural network intermediate results safety mechanism in an artificial neural network processor
US11874900B2 (en) 2020-09-29 2024-01-16 Hailo Technologies Ltd. Cluster interlayer safety mechanism in an artificial neural network processor
JPWO2022185527A1 (pt) * 2021-03-05 2022-09-09

Family Cites Families (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4212676B2 (ja) * 1998-02-12 2009-01-21 富士ゼロックス株式会社 情報処理システムおよび情報処理方法
US20080228719A1 (en) * 2007-03-13 2008-09-18 Fatdoor, Inc. People and business search result optimization
JP2008235992A (ja) * 2007-03-16 2008-10-02 Matsushita Electric Ind Co Ltd リコンフィギュラブル回路、リコンフィギュラブル回路システムおよびリコンフィギュラブル回路の配置配線方法
WO2010061911A1 (ja) * 2008-11-28 2010-06-03 日本電気株式会社 再構成可能回路生成装置、方法およびプログラム
US8442927B2 (en) * 2009-07-30 2013-05-14 Nec Laboratories America, Inc. Dynamically configurable, multi-ported co-processor for convolutional neural networks
CN102799563B (zh) * 2011-05-26 2015-08-26 上海红神信息技术有限公司 一种可重构计算阵列及构建方法
US10860946B2 (en) * 2011-08-10 2020-12-08 Konlanbi Dynamic data structures for data-driven modeling
US20130132962A1 (en) * 2011-11-22 2013-05-23 Microsoft Corporation Scheduler combinators
GB2516627A (en) * 2013-07-26 2015-02-04 Warsaw University Of Technology Parallel decision-or regression-tree growing
US10042894B2 (en) * 2013-10-31 2018-08-07 Microsoft Technology Licensing, Llc Temporal-based professional similarity
US9978014B2 (en) * 2013-12-18 2018-05-22 Intel Corporation Reconfigurable processing unit
CN104077233B (zh) * 2014-06-18 2017-04-05 百度在线网络技术（北京）有限公司 多通道卷积层处理方法和装置
CN104035751B (zh) 2014-06-20 2016-10-12 深圳市腾讯计算机系统有限公司 基于多图形处理器的数据并行处理方法及装置
JP6381368B2 (ja) * 2014-08-26 2018-08-29 キヤノン株式会社 画像処理装置、画像処理方法、およびプログラム
CN104287764B (zh) * 2014-09-11 2017-05-31 沈阳东软医疗系统有限公司 一种ct灌注成像方法和设备
US20160314410A1 (en) * 2015-04-23 2016-10-27 Rovi Guides, Inc. Systems and methods for improving accuracy in media asset recommendations based on data from one data space
CN104915195B (zh) * 2015-05-20 2017-11-28 清华大学 一种基于现场可编程门阵列实现神经网络计算的方法
CN104915322B (zh) * 2015-06-09 2018-05-01 中国人民解放军国防科学技术大学 一种卷积神经网络硬件加速方法
US10579628B2 (en) * 2015-12-17 2020-03-03 The Nielsen Company (Us), Llc Media names matching and normalization
CN105678379B (zh) * 2016-01-12 2020-08-07 腾讯科技（深圳）有限公司 一种cnn的处理方法和装置
US9903733B2 (en) * 2016-03-17 2018-02-27 Honda Motor Co., Ltd. Vehicular communications network and methods of use and manufacture thereof
CN106709532B (zh) * 2017-01-25 2020-03-10 京东方科技集团股份有限公司 图像处理方法和装置

Also Published As

Publication number Publication date
CN105678379A (zh) 2016-06-15
EP3404587A4 (en) 2019-10-30
US11151442B2 (en) 2021-10-19
CN105678379B (zh) 2020-08-07
WO2017121297A1 (zh) 2017-07-20
MY188759A (en) 2021-12-29
EP3404587B1 (en) 2023-11-01
EP3404587A1 (en) 2018-11-21
SG11201709689PA (en) 2017-12-28
KR20180005241A (ko) 2018-01-15
CA2987325C (en) 2019-10-22
US20180082175A1 (en) 2018-03-22
CA2987325A1 (en) 2017-07-20
JP6507271B2 (ja) 2019-04-24
JP2018526714A (ja) 2018-09-13
KR102192468B1 (ko) 2020-12-17

Similar Documents

Publication Publication Date Title
BR112017026917A2 (pt) ?método e dispositivo para processamento de rede neural convolucional cnn e mídia de armazenamento não volátil?
BR112015017722A8 (pt) Método e sistema de rastreio paralelo para desempenho e detalhe
CY1124626T1 (el) Υπολογιστικο συστημα και μεθοδος υψηλων επιδοσεων
BR112016002637A8 (pt) Sincronização de barreira com cálculo dinâmico de largura
BR112018002040A2 (pt) controle de uma nuvem de dispositivo
IN2013MU03644A (pt)
BR112019013067B8 (pt) Método e dispositivo para processamento de serviço
BR112018004933A2 (pt) agendamento eficaz de tarefas multiversionadas
BR112018007623A2 (pt) método para programação eficiente de tarefa na presença de conflitos
BR112015003406A8 (pt) Método implementado por computador e sistema de computação
GB2510769A (en) Processing element management in a streaming data system
AR087681A1 (es) Sistema y metodo para proveer acceso a internet a un dispositivo informatico
BR112019004924A2 (pt) coerência dinâmica de entrada/saída
BR112018005970A2 (pt) mecanismo de reparo de marca de tempo no caso de falha de descompressão
JP2016509714A5 (pt)
BR112018008737A8 (pt) método para fornecer módulos de filtro, produto de programa de computador, e aparelho para gerenciamento de processo
BR112018006349A2 (pt) preempção de unidade de processamento gráfico com granularidade de nível de bloco de pixel
BR112015017859A8 (pt) Método e dispositivo para agrupar imagens
SG10201901587VA (en) Application testing
BR112017004299A2 (pt) dispositivo e método de processamento de informação, e, programa.
BR112019003512A2 (pt) método de processamento de função de rede e dispositivos relacionados
BR112015016012A2 (pt) método e dispositivo para transferência de recursos
BR112018077404A2 (pt) sistema e método de marcação em grupo baseados na aprendizagem
BR112018001293A2 (pt) dispositivo de computação óptico e método
BR112013004781A2 (pt) dispositivo de processamento de imagem em movimento, método de processamento de imagem em movimento, e, programa

Legal Events

Date Code Title Description
B350 Update of information on the portal [chapter 15.35 patent gazette]
B06W Patent application suspended after preliminary examination (for patents with searches from other patent authorities) chapter 6.23 patent gazette]
B15K Others concerning applications: alteration of classification

Free format text: A CLASSIFICACAO ANTERIOR ERA: G06N 3/06

Ipc: G06N 3/063 (2023.01), G06N 3/045 (2023.01)

B09A Decision: intention to grant [chapter 9.1 patent gazette]
B09W Correction of the decision to grant [chapter 9.1.4 patent gazette]

Free format text: DEVIDO A INCORRECOES NO QUADRO 01.