CN110506280A - 神经网络训练系统、方法和计算机可读存储介质 - Google Patents

神经网络训练系统、方法和计算机可读存储介质 Download PDF

Info

Publication number
CN110506280A
CN110506280A CN201880025109.7A CN201880025109A CN110506280A CN 110506280 A CN110506280 A CN 110506280A CN 201880025109 A CN201880025109 A CN 201880025109A CN 110506280 A CN110506280 A CN 110506280A
Authority
CN
China
Prior art keywords
processing device
vector
neural network
weight vector
correction value
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.)
Granted
Application number
CN201880025109.7A
Other languages
English (en)
Other versions
CN110506280B (zh
Inventor
费旭东
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.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
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 Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Publication of CN110506280A publication Critical patent/CN110506280A/zh
Application granted granted Critical
Publication of CN110506280B publication Critical patent/CN110506280B/zh
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • G06N3/08—Learning methods
    • 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
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Computational Linguistics (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Neurology (AREA)
  • Image Analysis (AREA)
  • Feedback Control In General (AREA)
  • Manipulator (AREA)

Abstract

一种神经网络训练系统、方法和计算机可读存储介质，属于机器学习算法领域。所述神经网络训练系统(300)包括：第一处理设备(301)和第二处理设备(302)，其中，第一处理设备(301)用于获取目标神经网络的权向量和训练集中的N个数据向量，并基于N个数据向量中的每一个数据向量和权向量进行第一运算得到N组输出值，第一运算包括向量点积运算；第二处理设备(302)，用于获取根据N组输出值计算得到的至少一个修正值，并根据至少一个修正值对第二处理设备(302)中存储的神经网络的权向量中的向量元素进行修正，得到修正权向量，并将修正权向量发送至第一处理设备(301)。提供的神经网络训练系统能够提高神经网络训练的效率。

Description

PCT国内申请，说明书已公开。

Claims (23)

  1. PCT国内申请，权利要求书已公开。
CN201880025109.7A 2017-08-22 2018-03-19 神经网络训练系统、方法和计算机可读存储介质 Active CN110506280B (zh)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
CN2017107257759 2017-08-22
CN201710725775.9A CN109426859B (zh) 2017-08-22 2017-08-22 神经网络训练系统、方法和计算机可读存储介质
PCT/CN2018/079500 WO2019037409A1 (zh) 2017-08-22 2018-03-19 神经网络训练系统、方法和计算机可读存储介质

Publications (2)

Publication Number Publication Date
CN110506280A true CN110506280A (zh) 2019-11-26
CN110506280B CN110506280B (zh) 2022-12-27

Family

ID=65438345

Family Applications (2)

Application Number Title Priority Date Filing Date
CN201710725775.9A Active CN109426859B (zh) 2017-08-22 2017-08-22 神经网络训练系统、方法和计算机可读存储介质
CN201880025109.7A Active CN110506280B (zh) 2017-08-22 2018-03-19 神经网络训练系统、方法和计算机可读存储介质

Family Applications Before (1)

Application Number Title Priority Date Filing Date
CN201710725775.9A Active CN109426859B (zh) 2017-08-22 2017-08-22 神经网络训练系统、方法和计算机可读存储介质

Country Status (2)

Country Link
CN (2) CN109426859B (zh)
WO (1) WO2019037409A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109426859B (zh) * 2017-08-22 2021-03-05 华为技术有限公司 神经网络训练系统、方法和计算机可读存储介质
CN111783932A (zh) * 2019-04-03 2020-10-16 华为技术有限公司 训练神经网络的方法和装置
CN111126596B (zh) * 2019-12-17 2021-03-19 百度在线网络技术（北京）有限公司 神经网络训练中的信息处理方法、设备与存储介质
CN113177355B (zh) * 2021-04-28 2024-01-12 南方电网科学研究院有限责任公司 一种电力负荷预测方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1672171A (zh) * 2002-07-31 2005-09-21 雅马哈发动机株式会社 基于量子软计算的智能机电控制悬架系统
CN101101299A (zh) * 2007-06-25 2008-01-09 华东理工大学 一种并-串联模式识别方法及其在机器嗅觉中的应用
CN101449264A (zh) * 2006-07-12 2009-06-03 柯法克斯公司 用于转导数据分类的方法和系统以及使用机器学习方法的数据分类方法
CN103049792A (zh) * 2011-11-26 2013-04-17 微软公司 深层神经网络的辨别预训练
CN106203622A (zh) * 2016-07-14 2016-12-07 杭州华为数字技术有限公司 神经网络运算装置
CN107688493A (zh) * 2016-08-05 2018-02-13 阿里巴巴集团控股有限公司 训练深度神经网络的方法、装置及系统
CN109426859A (zh) * 2017-08-22 2019-03-05 华为技术有限公司 神经网络训练系统、方法和计算机可读存储介质

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
AU2009279644B2 (en) * 2008-08-06 2012-08-23 Halliburton Energy Services, Inc. Systems and methods employing cooperative optimization-based dimensionality reduction
EP3089081A4 (en) * 2014-02-10 2017-09-20 Mitsubishi Electric Corporation Hierarchical neural network device, learning method for determination device, and determination method
CN105654176B (zh) * 2014-11-14 2018-03-27 富士通株式会社 神经网络系统及神经网络系统的训练装置和方法
CN105678395B (zh) * 2014-11-21 2021-06-29 创新先进技术有限公司 神经网络的建立方法及系统和神经网络的应用方法及系统
CN106203616A (zh) * 2015-05-04 2016-12-07 富士通株式会社 神经网络模型训练装置和方法

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1672171A (zh) * 2002-07-31 2005-09-21 雅马哈发动机株式会社 基于量子软计算的智能机电控制悬架系统
CN101449264A (zh) * 2006-07-12 2009-06-03 柯法克斯公司 用于转导数据分类的方法和系统以及使用机器学习方法的数据分类方法
CN101101299A (zh) * 2007-06-25 2008-01-09 华东理工大学 一种并-串联模式识别方法及其在机器嗅觉中的应用
CN103049792A (zh) * 2011-11-26 2013-04-17 微软公司 深层神经网络的辨别预训练
CN106203622A (zh) * 2016-07-14 2016-12-07 杭州华为数字技术有限公司 神经网络运算装置
CN107688493A (zh) * 2016-08-05 2018-02-13 阿里巴巴集团控股有限公司 训练深度神经网络的方法、装置及系统
CN109426859A (zh) * 2017-08-22 2019-03-05 华为技术有限公司 神经网络训练系统、方法和计算机可读存储介质

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
蒋大明: "《自控遥控基础》", 30 April 1998 *

Also Published As

Publication number Publication date
CN110506280B (zh) 2022-12-27
CN109426859B (zh) 2021-03-05
WO2019037409A1 (zh) 2019-02-28
CN109426859A (zh) 2019-03-05

Similar Documents

Publication Publication Date Title
CN110506280B (zh) 神经网络训练系统、方法和计算机可读存储介质
Bock et al. An improvement of the convergence proof of the ADAM-Optimizer
KR102149132B1 (ko) 딥 러닝 환경에서의 적대적 예제 생성 장치 및 방법, 컴퓨터 프로그램
Son Finite-time stability of a class of non-autonomous neural networks with heterogeneous proportional delays
US20210334621A1 (en) Arithmetic processing system using hierarchical network
KR101885593B1 (ko) 인공 지능을 기반으로 하는 자소 음소 변환 모델 생성 방법 및 장치
CN109919183B (zh) 一种基于小样本的图像识别方法、装置、设备及存储介质
Al-Faiz et al. The effect of Z-Score standardization (normalization) on binary input due the speed of learning in back-propagation neural network
GB2604271A (en) Master transform architecture for deep learning
WO2020151310A1 (zh) 文本生成方法、装置、计算机设备及介质
KR20210116923A (ko) 디노이징 네트워크의 학습 방법, 및 이미지 처리 방법 및 그 장치 방법
Lan et al. Random search enhancement of error minimized extreme learning machine.
CN108304912B (zh) 一种运用抑制信号实现脉冲神经网络监督学习的系统和方法
WO2017166155A1 (zh) 一种对神经网络模型进行训练的方法、装置及电子设备
Mushgil et al. Comparison between resilient and standard back propagation algorithms efficiency in pattern recognition
WO2019155523A1 (ja) 分類器形成装置、分類器形成方法、及びプログラムを格納する非一時的なコンピュータ可読媒体
CN101169621A (zh) 一种基于小波神经网络的视觉反馈控制方法
KR20210064817A (ko) 상이한 딥러닝 모델 간의 전이 학습방법
Aizenberg et al. Batch linear least squares-based learning algorithm for MLMVN with soft margins
Ito et al. Generalization in learning multiple temporal patterns using rnnpb
Yonekawa et al. The parameter optimization of the pulse coupled neural network for the pattern recognition
KR102101481B1 (ko) 인공지능 기반의 휴대용 보안영상 학습장치 및 방법
CN110163231B (zh) 建立影像数据识别模型的方法、计算机系统及记录介质
JP2021197032A (ja) 制御装置、方法及びプログラム
Zhang et al. A novel noise injection-based training scheme for better model robustness

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant