AU2016256315A1 - Incorporating top-down information in deep neural networks via the bias term - Google Patents

Incorporating top-down information in deep neural networks via the bias term Download PDF

Info

Publication number
AU2016256315A1
AU2016256315A1 AU2016256315A AU2016256315A AU2016256315A1 AU 2016256315 A1 AU2016256315 A1 AU 2016256315A1 AU 2016256315 A AU2016256315 A AU 2016256315A AU 2016256315 A AU2016256315 A AU 2016256315A AU 2016256315 A1 AU2016256315 A1 AU 2016256315A1
Authority
AU
Australia
Prior art keywords
bias
network
input
adjusting
present
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.)
Abandoned
Application number
AU2016256315A
Other languages
English (en)
Inventor
Regan Blythe TOWAL
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.)
Qualcomm Inc
Original Assignee
Qualcomm 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 Qualcomm Inc filed Critical Qualcomm Inc
Publication of AU2016256315A1 publication Critical patent/AU2016256315A1/en
Abandoned legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06N3/088—Non-supervised learning, e.g. competitive learning
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06N3/048—Activation functions
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/01—Probabilistic graphical models, e.g. probabilistic networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Probability & Statistics with Applications (AREA)
  • Algebra (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)
  • User Interface Of Digital Computer (AREA)
AU2016256315A 2015-04-28 2016-03-11 Incorporating top-down information in deep neural networks via the bias term Abandoned AU2016256315A1 (en)

Applications Claiming Priority (5)

Application Number Priority Date Filing Date Title
US201562154097P 2015-04-28 2015-04-28
US62/154,097 2015-04-28
US14/848,288 2015-09-08
US14/848,288 US10325202B2 (en) 2015-04-28 2015-09-08 Incorporating top-down information in deep neural networks via the bias term
PCT/US2016/022158 WO2016175925A1 (en) 2015-04-28 2016-03-11 Incorporating top-down information in deep neural networks via the bias term

Publications (1)

Publication Number Publication Date
AU2016256315A1 true AU2016256315A1 (en) 2017-10-05

Family

ID=55586459

Family Applications (1)

Application Number Title Priority Date Filing Date
AU2016256315A Abandoned AU2016256315A1 (en) 2015-04-28 2016-03-11 Incorporating top-down information in deep neural networks via the bias term

Country Status (8)

Country Link
US (1) US10325202B2 (enExample)
EP (1) EP3289527A1 (enExample)
JP (1) JP2018518740A (enExample)
KR (1) KR20170140228A (enExample)
CN (1) CN107533665A (enExample)
AU (1) AU2016256315A1 (enExample)
BR (1) BR112017022983A2 (enExample)
WO (1) WO2016175925A1 (enExample)

Families Citing this family (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR3044438A1 (fr) * 2015-11-27 2017-06-02 Thales Sa Systeme et procede d'aide a la decision
WO2018035805A1 (en) * 2016-08-25 2018-03-01 Intel Corporation Coupled multi-task fully convolutional networks using multi-scale contextual information and hierarchical hyper-features for semantic image segmentation
CA3051990C (en) * 2017-02-23 2021-03-23 Cerebras Systems Inc. Accelerated deep learning
US11842280B2 (en) 2017-05-05 2023-12-12 Nvidia Corporation Loss-scaling for deep neural network training with reduced precision
WO2018218651A1 (en) 2017-06-02 2018-12-06 Nokia Technologies Oy Artificial neural network
US10108538B1 (en) 2017-07-31 2018-10-23 Google Llc Accessing prologue and epilogue data
US11144815B2 (en) * 2017-12-04 2021-10-12 Optimum Semiconductor Technologies Inc. System and architecture of neural network accelerator
KR102153791B1 (ko) * 2017-12-20 2020-09-08 연세대학교 산학협력단 인공 신경망을 위한 디지털 뉴런, 인공 뉴런 및 이를 포함하는 추론 엔진
US11531930B2 (en) * 2018-03-12 2022-12-20 Royal Bank Of Canada System and method for monitoring machine learning models
CN108977897B (zh) * 2018-06-07 2021-11-19 浙江天悟智能技术有限公司 基于局部内在可塑性回声状态网络的熔纺工艺控制方法
US20210350236A1 (en) * 2018-09-28 2021-11-11 National Technology & Engineering Solutions Of Sandia, Llc Neural network robustness via binary activation
KR102184655B1 (ko) * 2018-10-29 2020-11-30 에스케이텔레콤 주식회사 비대칭 tanh 활성 함수를 이용한 예측 성능의 개선
US11481667B2 (en) * 2019-01-24 2022-10-25 International Business Machines Corporation Classifier confidence as a means for identifying data drift
US20200242771A1 (en) * 2019-01-25 2020-07-30 Nvidia Corporation Semantic image synthesis for generating substantially photorealistic images using neural networks
DE102019217444A1 (de) * 2019-11-12 2021-05-12 Robert Bosch Gmbh Verfahren und Vorrichtung zur Klassifizierung digitaler Bilddaten
US10929748B1 (en) * 2019-11-26 2021-02-23 Mythic, Inc. Systems and methods for implementing operational transformations for restricted computations of a mixed-signal integrated circuit
US12154032B2 (en) * 2020-02-04 2024-11-26 Dsp Group Ltd. Post-training control of the bias of neural networks
KR20210158697A (ko) * 2020-06-24 2021-12-31 삼성전자주식회사 뉴로모픽 장치 및 뉴로모픽 장치를 이용하여 뉴럴 네트워크를 구현하는 방법
US12216740B2 (en) 2021-01-08 2025-02-04 Bank Of America Corporation Data source evaluation platform for improved generation of supervised learning models
CN113473580B (zh) * 2021-05-14 2024-04-26 南京信息工程大学滨江学院 异构网络中基于深度学习的用户关联联合功率分配方法

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000112914A (ja) * 1998-10-09 2000-04-21 Toshiba Corp 神経回路網の学習方法
WO2008066731A2 (en) 2006-11-22 2008-06-05 Psigenics Corporation Device and method responsive to influences of mind
WO2010030794A1 (en) 2008-09-10 2010-03-18 Digital Infuzion, Inc. Machine learning methods and systems for identifying patterns in data
US8239336B2 (en) 2009-03-09 2012-08-07 Microsoft Corporation Data processing using restricted boltzmann machines
US8754802B2 (en) 2010-08-26 2014-06-17 Lawrence Livermore National Security, Llc Determining root correspondence between previously and newly detected objects
US20150019468A1 (en) 2013-07-09 2015-01-15 Knowmtech, Llc Thermodynamic computing
CN103778414A (zh) * 2014-01-17 2014-05-07 杭州电子科技大学 基于深度神经网络的实时人脸识别方法
CN104200224A (zh) * 2014-08-28 2014-12-10 西北工业大学 基于深度卷积神经网络的无价值图像去除方法

Also Published As

Publication number Publication date
WO2016175925A1 (en) 2016-11-03
US10325202B2 (en) 2019-06-18
US20160321542A1 (en) 2016-11-03
JP2018518740A (ja) 2018-07-12
KR20170140228A (ko) 2017-12-20
EP3289527A1 (en) 2018-03-07
BR112017022983A2 (pt) 2018-07-24
CN107533665A (zh) 2018-01-02

Similar Documents

Publication Publication Date Title
US10325202B2 (en) Incorporating top-down information in deep neural networks via the bias term
EP3332362B1 (en) Detection of unknown classes and initialization of classifiers for unknown classes
US10318848B2 (en) Methods for object localization and image classification
US10878320B2 (en) Transfer learning in neural networks
US10515304B2 (en) Filter specificity as training criterion for neural networks
US10275719B2 (en) Hyper-parameter selection for deep convolutional networks
US11334789B2 (en) Feature selection for retraining classifiers
US10410096B2 (en) Context-based priors for object detection in images
US11423323B2 (en) Generating a sparse feature vector for classification
US20160283864A1 (en) Sequential image sampling and storage of fine-tuned features
US20170091619A1 (en) Selective backpropagation
US20170032247A1 (en) Media classification

Legal Events

Date Code Title Description
MK1 Application lapsed section 142(2)(a) - no request for examination in relevant period