WO2020113240A3 - Methods and apparatus for training optical deep diffractive neural networks - Google Patents

Methods and apparatus for training optical deep diffractive neural networks Download PDF

Info

Publication number
WO2020113240A3
WO2020113240A3 PCT/US2019/064701 US2019064701W WO2020113240A3 WO 2020113240 A3 WO2020113240 A3 WO 2020113240A3 US 2019064701 W US2019064701 W US 2019064701W WO 2020113240 A3 WO2020113240 A3 WO 2020113240A3
Authority
WO
WIPO (PCT)
Prior art keywords
training
light
methods
neural networks
information
Prior art date
Application number
PCT/US2019/064701
Other languages
French (fr)
Other versions
WO2020113240A2 (en
Inventor
Yongxi Tan
Jin Yang
Yimeng Li
Xiang Liu
Ning Cheng
Huaxia WANG
Original Assignee
Futurewei Technologies, Inc.
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 Futurewei Technologies, Inc. filed Critical Futurewei Technologies, Inc.
Priority to PCT/US2019/064701 priority Critical patent/WO2020113240A2/en
Publication of WO2020113240A2 publication Critical patent/WO2020113240A2/en
Publication of WO2020113240A3 publication Critical patent/WO2020113240A3/en

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
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • 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
    • G06N3/06—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
    • 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
    • G06N3/10—Interfaces, programming languages or software development kits, e.g. for simulating 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
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield 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
    • G06N3/047—Probabilistic or stochastic networks

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)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Molecular Biology (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • Neurology (AREA)
  • Eye Examination Apparatus (AREA)
  • Optical Modulation, Optical Deflection, Nonlinear Optics, Optical Demodulation, Optical Logic Elements (AREA)

Abstract

A method for training a neural network (NN) includes receiving partition information of a light element partition of a spatially incoherent light source (SILS) and training information, the light element partition being assigned to the training processor; modulating the individual light elements in the light element partition in accordance with the training information; detecting diffracted light from a diffractive deep neural network (D2NN) of the NN; and sending information corresponding to the detected diffracted light to the controller.
PCT/US2019/064701 2019-12-05 2019-12-05 Methods and apparatus for training optical deep diffractive neural networks WO2020113240A2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/US2019/064701 WO2020113240A2 (en) 2019-12-05 2019-12-05 Methods and apparatus for training optical deep diffractive neural networks

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/US2019/064701 WO2020113240A2 (en) 2019-12-05 2019-12-05 Methods and apparatus for training optical deep diffractive neural networks

Publications (2)

Publication Number Publication Date
WO2020113240A2 WO2020113240A2 (en) 2020-06-04
WO2020113240A3 true WO2020113240A3 (en) 2020-10-01

Family

ID=69006059

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2019/064701 WO2020113240A2 (en) 2019-12-05 2019-12-05 Methods and apparatus for training optical deep diffractive neural networks

Country Status (1)

Country Link
WO (1) WO2020113240A2 (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022056422A1 (en) * 2020-09-14 2022-03-17 The Regents Of The University Of California Ensemble learning of diffractive neural networks
CN112418403B (en) * 2020-11-25 2022-06-28 清华大学 Optical diffraction computing processor based on optical diffraction principle and programmable device

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180322387A1 (en) * 2017-05-05 2018-11-08 Intel Corporation Hardware implemented point to point communication primitives for machine learning
WO2019200289A1 (en) * 2018-04-13 2019-10-17 The Regents Of The University Of California Devices and methods employing optical-based machine learning using diffractive deep neural networks

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180322387A1 (en) * 2017-05-05 2018-11-08 Intel Corporation Hardware implemented point to point communication primitives for machine learning
WO2019200289A1 (en) * 2018-04-13 2019-10-17 The Regents Of The University Of California Devices and methods employing optical-based machine learning using diffractive deep neural networks

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
JULIE CHANG ET AL: "Hybrid optical-electronic convolutional neural networks with optimized diffractive optics for image classification", SCIENTIFIC REPORTS, vol. 8, no. 1, 17 August 2018 (2018-08-17), XP055602493, DOI: 10.1038/s41598-018-30619-y *
SHUMING JIAO ET AL: "Optical machine learning with incoherent light and a single-pixel detector", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 24 April 2019 (2019-04-24), XP081537995, DOI: 10.1364/OL.44.005186 *

Also Published As

Publication number Publication date
WO2020113240A2 (en) 2020-06-04

Similar Documents

Publication Publication Date Title
WO2017052709A3 (en) Transfer learning in neural networks
WO2020132102A3 (en) Neural networks for coarse- and fine-object classifications
WO2020008192A3 (en) Phenotype prediction
JOP20220067A1 (en) System to collect and identify skin conditions from images and expert knowledge
WO2020113240A3 (en) Methods and apparatus for training optical deep diffractive neural networks
WO2018146558A3 (en) Systems, methods and apparatuses for stereo vision and tracking
EP3903247A4 (en) Method, apparatus and system for secure vertical federated learning
EP3101597A3 (en) Reading comprehension neural networks
EP4035096A4 (en) Decentralised artificial intelligence (ai)/machine learning training system
MX2019000713A (en) Distributed machine learning systems, apparatus, and methods.
EP3834166A4 (en) Cloud-based, data-driven artificial intelligence and machine learning financial planning and analysis visualization platform
NZ722904A (en) Multi-focal display system and method
WO2015199789A3 (en) Optical based pose detection for multiple unmanned underwater vehicles
WO2014115129A3 (en) Dynamic aperture holography
NO20181065A1 (en) Optical imaging and assessment system for tong cassette positioning device
EP3887922A4 (en) Data center infrastructure optimization method based on causal learning
EP3737985A4 (en) Varifocal apparatuses, systems, and methods employing a deformable stepped lens
EP4036561A4 (en) Data generation system, learning device, data generation device, data generation method, and data generation program
WO2020101863A3 (en) Hybrid opto-electronic neural networks and methods for utilization thereof
EP3936224A4 (en) Data generation device, data generation method, learning device, and learning method
EP3990967A4 (en) Apparatus and methods for eye tracking based on eye imaging via a light-guide optical element
EP3352426A3 (en) Optimizing network efficiency for application requirements
EP3860001A4 (en) Single board, optical module, olt, and information processing method
EP3767582A4 (en) Inspection system, discrimination system, and learning data generator
EP2916237A3 (en) A system and method for rectifying a typographical error in a text file

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19828129

Country of ref document: EP

Kind code of ref document: A2

122 Ep: pct application non-entry in european phase

Ref document number: 19828129

Country of ref document: EP

Kind code of ref document: A2