CA2095720A1 - Method and apparatus for training a neural network - Google Patents

Method and apparatus for training a neural network

Info

Publication number
CA2095720A1
CA2095720A1 CA 2095720 CA2095720A CA2095720A1 CA 2095720 A1 CA2095720 A1 CA 2095720A1 CA 2095720 CA2095720 CA 2095720 CA 2095720 A CA2095720 A CA 2095720A CA 2095720 A1 CA2095720 A1 CA 2095720A1
Authority
CA
Canada
Prior art keywords
vectors
network
parameters
similarity
controller
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
CA 2095720
Other languages
French (fr)
Other versions
CA2095720C (en
Inventor
David Charles Wood
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.)
British Telecommunications PLC
Original Assignee
Individual
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 Individual filed Critical Individual
Publication of CA2095720A1 publication Critical patent/CA2095720A1/en
Application granted granted Critical
Publication of CA2095720C publication Critical patent/CA2095720C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

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
    • G06N3/067—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using optical means
    • G06N3/0675—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using optical means using electro-optical, acousto-optical or opto-electronic means
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B82—NANOTECHNOLOGY
    • B82Y—SPECIFIC USES OR APPLICATIONS OF NANOSTRUCTURES; MEASUREMENT OR ANALYSIS OF NANOSTRUCTURES; MANUFACTURE OR TREATMENT OF NANOSTRUCTURES
    • B82Y20/00—Nanooptics, e.g. quantum optics or photonic crystals

Abstract

A method of training a neural network (2) having dynamically adjustable parameters controlled by a controller (10) which determine the response of the network (2). A set of input vectors (I l to I n) are input to network (2) at an input port (4). The corresponding set of output vectors (O'l to O'n) provided by the network (2) are compared to a target set of output vectors (O l to O n) by an error logger (12) which provides to the controller (10) a measure of similarity of the two sets. The controller (10) is arranged to alter the dynamic parameters independence on the average number of occasions the output vectors are different from the respective target output vectors. Measuring the similarity of the whole of the output set and target set and adjusting the parameters on this global measure rather than on the similarity of pairs of individual vectors provides enhanced training rates for neural networks having a data throughput rate that can be higher than the rate at which the parameters can be adjusted.
CA002095720A 1990-11-08 1991-11-08 Method and apparatus for training a neural network Expired - Fee Related CA2095720C (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
GB909024332A GB9024332D0 (en) 1990-11-08 1990-11-08 Method of training a neural network
GB9024332.0 1990-11-08
PCT/GB1991/001967 WO1992009044A1 (en) 1990-11-08 1991-11-08 Method of training a neural network

Publications (2)

Publication Number Publication Date
CA2095720A1 true CA2095720A1 (en) 1992-05-09
CA2095720C CA2095720C (en) 1999-02-16

Family

ID=10685082

Family Applications (1)

Application Number Title Priority Date Filing Date
CA002095720A Expired - Fee Related CA2095720C (en) 1990-11-08 1991-11-08 Method and apparatus for training a neural network

Country Status (7)

Country Link
US (1) US5390285A (en)
EP (1) EP0556254B1 (en)
JP (1) JP3290984B2 (en)
CA (1) CA2095720C (en)
DE (1) DE69119604T2 (en)
GB (1) GB9024332D0 (en)
WO (1) WO1992009044A1 (en)

Families Citing this family (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE69428495T2 (en) * 1993-11-26 2002-04-11 Koninkl Philips Electronics Nv Multimode laser for an optical information processing system, in particular for a neural network
US5903884A (en) * 1995-08-08 1999-05-11 Apple Computer, Inc. Method for training a statistical classifier with reduced tendency for overfitting
KR100267839B1 (en) 1995-11-06 2000-10-16 오가와 에이지 Nitride semiconductor device
US5872975A (en) * 1996-06-05 1999-02-16 Lockheed Martin Corporation Automatic retargeting of processor modules in multiple processor systems
US6501857B1 (en) * 1999-07-20 2002-12-31 Craig Gotsman Method and system for detecting and classifying objects in an image
US6424960B1 (en) * 1999-10-14 2002-07-23 The Salk Institute For Biological Studies Unsupervised adaptation and classification of multiple classes and sources in blind signal separation
US20050149462A1 (en) * 1999-10-14 2005-07-07 The Salk Institute For Biological Studies System and method of separating signals
US7130776B2 (en) * 2002-03-25 2006-10-31 Lockheed Martin Corporation Method and computer program product for producing a pattern recognition training set
US7082420B2 (en) * 2002-07-13 2006-07-25 James Ting-Ho Lo Convexification method of training neural networks and estimating regression models
US11222263B2 (en) 2016-07-28 2022-01-11 Samsung Electronics Co., Ltd. Neural network method and apparatus
US11244226B2 (en) 2017-06-12 2022-02-08 Nvidia Corporation Systems and methods for training neural networks with sparse data
US10565686B2 (en) 2017-06-12 2020-02-18 Nvidia Corporation Systems and methods for training neural networks for regression without ground truth training samples
JP6838259B2 (en) * 2017-11-08 2021-03-03 Kddi株式会社 Learning data generator, judgment device and program
CN110956259B (en) * 2019-11-22 2023-05-12 联合微电子中心有限责任公司 Photon neural network training method based on forward propagation
CN111722923A (en) * 2020-05-29 2020-09-29 浪潮电子信息产业股份有限公司 Heterogeneous resource calling method and device and computer readable storage medium
CN112447188B (en) * 2020-11-18 2023-10-20 中国人民解放军陆军工程大学 Acoustic scene classification method based on improved softmax function

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4918618A (en) * 1988-04-11 1990-04-17 Analog Intelligence Corporation Discrete weight neural network
US5099434A (en) * 1988-07-18 1992-03-24 Northrop Corporation Continuous-time optical neural network
US5004309A (en) * 1988-08-18 1991-04-02 Teledyne Brown Engineering Neural processor with holographic optical paths and nonlinear operating means
US4914603A (en) * 1988-12-14 1990-04-03 Gte Laboratories Incorporated Training neural networks
US5068801A (en) * 1989-11-06 1991-11-26 Teledyne Industries, Inc. Optical interconnector and highly interconnected, learning neural network incorporating optical interconnector therein

Also Published As

Publication number Publication date
JP3290984B2 (en) 2002-06-10
CA2095720C (en) 1999-02-16
JPH06504636A (en) 1994-05-26
DE69119604D1 (en) 1996-06-20
WO1992009044A1 (en) 1992-05-29
EP0556254B1 (en) 1996-05-15
GB9024332D0 (en) 1990-12-19
EP0556254A1 (en) 1993-08-25
US5390285A (en) 1995-02-14
DE69119604T2 (en) 1996-09-26

Similar Documents

Publication Publication Date Title
CA2095720A1 (en) Method and apparatus for training a neural network
CA2066278A1 (en) Computer neural network supervisory process control system and method
US5649063A (en) Feedback process control using a neural network parameter estimator
CA2165229A1 (en) Method and Apparatus for Characterizing an Input Signal
ATE111289T1 (en) MEASUREMENT OF HEARING AID USING VECTORS.
US5598329A (en) Method and device for controlling a process
EP0362840A3 (en) Neural network system
FR2454238A1 (en) ADJUSTING DEVICE FOR COLOR TELEVISION APPARATUS
EP0377908A3 (en) Neural network having an associative memory that learns by example
DE69821582D1 (en) Device for the production of band-shaped edible dough
SE8804196L (en) PROCEDURE AND DEVICE TO RESTORE A DATA SIGNAL
Fang et al. Faster learning for dynamic recurrent backpropagation.
DE69516193D1 (en) Image processing method and device
DK0594885T3 (en) Process for regulating a heating system and apparatus for carrying out the method
US5297075A (en) Computer controlled transversal equalizer
DE69030812T2 (en) METHOD AND DEVICE FOR CONTROLLING A PROCESS WITH THE AID OF A TRAINED PARALLEL DISTRIBUTED PROCESSING NETWORK
EP0539010A3 (en) Method and device for generating sum information/rounding control signal
Christensen et al. The auditory tau effect and memory for pitch
SU1124883A3 (en) Arrangement for controlling geometrical dimensions of rolled product on continuous rolling mill
DK170682A (en) PROCEDURE AND APPARATUS FOR REGULATING A HEATING FLUID
CA2065984A1 (en) Exposure control apparatus of camera
SU879608A1 (en) Device for simulating neuron
Sujaku et al. Binaural interaction in the cat inferior colliculus: comparison of the physiological data with a computer simulated model
JPH06243273A (en) Learning system for neural network
JPS6465432A (en) Noninterference control system for testing device on engine board

Legal Events

Date Code Title Description
EEER Examination request
MKLA Lapsed