WO2020159638A1 - Système et procédé d'adaptation de domaine non supervisé par l'intermédiaire d'une distance de wasserstein segmentée - Google Patents

Système et procédé d'adaptation de domaine non supervisé par l'intermédiaire d'une distance de wasserstein segmentée Download PDF

Info

Publication number
WO2020159638A1
WO2020159638A1 PCT/US2019/067259 US2019067259W WO2020159638A1 WO 2020159638 A1 WO2020159638 A1 WO 2020159638A1 US 2019067259 W US2019067259 W US 2019067259W WO 2020159638 A1 WO2020159638 A1 WO 2020159638A1
Authority
WO
WIPO (PCT)
Prior art keywords
input data
data distribution
data
labels
task
Prior art date
Application number
PCT/US2019/067259
Other languages
English (en)
Inventor
Alexander J. GABOURIE
Mohammad Rostami
Soheil KOLOURI
Kyungnam Kim
Original Assignee
Hrl Laboratories, Llc
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
Priority claimed from US16/262,878 external-priority patent/US11620527B2/en
Application filed by Hrl Laboratories, Llc filed Critical Hrl Laboratories, Llc
Priority to EP19842474.9A priority Critical patent/EP3918527A1/fr
Priority to CN201980087199.7A priority patent/CN113316790B/zh
Publication of WO2020159638A1 publication Critical patent/WO2020159638A1/fr

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
    • 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
    • G06N3/045—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
    • G06N3/047—Probabilistic or stochastic networks

Definitions

  • the input device 112 is coupled with the address/data bus 102, wherein the input device 112 is configured to communicate information and command selections to the processor 100.
  • the input device 112 is an alphanumeric input device, such as a keyboard, that may include alphanumeric and/or function keys.
  • a storage device 116 coupled with the address/data bus 102.
  • the storage device 116 is configured to store information and/or computer executable instructions.
  • the storage device 116 is a storage device such as a magnetic or optical disk drive (e.g., hard disk drive (“HDD”), floppy diskette, compact disk read only memory (“CD-ROM”), digital versatile disk (“DVD”)).
  • a display device 118 is coupled with the address/data bus 102, wherein the display device 118 is configured to display video and/or graphics.
  • Other appropriate responses may include one or more of a steering operation, a throttle operation to increase speed or to decrease speed, or a decision to maintain course and speed without change.
  • the responses may be appropriate for avoiding a collision, improving travel speed, or improving efficiency.
  • control of other device types is also possible.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Molecular Biology (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Probability & Statistics with Applications (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Image Analysis (AREA)

Abstract

L'invention concerne un système d'adaptation de domaine non supervisé dans un agent d'apprentissage autonome. Le système adapte un modèle appris avec un ensemble de données non étiquetées à partir d'un domaine cible, ce qui permet d'obtenir un modèle adapté. Le modèle appris a été préalablement formé pour effectuer une tâche à l'aide d'un ensemble de données étiquetées à partir d'un domaine source. L'ensemble de données étiquetées a une première distribution de données d'entrée, et l'ensemble de données cibles non étiquetées a une seconde distribution de données d'entrée qui est distincte de la première distribution de données d'entrée. Le modèle adapté est mis en œuvre dans l'agent d'apprentissage autonome, amenant l'agent d'apprentissage autonome à effectuer la tâche dans le domaine cible.
PCT/US2019/067259 2019-01-30 2019-12-18 Système et procédé d'adaptation de domaine non supervisé par l'intermédiaire d'une distance de wasserstein segmentée WO2020159638A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP19842474.9A EP3918527A1 (fr) 2019-01-30 2019-12-18 Système et procédé d'adaptation de domaine non supervisé par l'intermédiaire d'une distance de wasserstein segmentée
CN201980087199.7A CN113316790B (zh) 2019-01-30 2019-12-18 用于自主学习代理中的无监督域适应的系统、方法和介质

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US16/262,878 US11620527B2 (en) 2018-02-06 2019-01-30 Domain adaption learning system
US16/262,878 2019-01-30
US201962807716P 2019-02-19 2019-02-19
US62/807,716 2019-02-19

Publications (1)

Publication Number Publication Date
WO2020159638A1 true WO2020159638A1 (fr) 2020-08-06

Family

ID=71841627

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2019/067259 WO2020159638A1 (fr) 2019-01-30 2019-12-18 Système et procédé d'adaptation de domaine non supervisé par l'intermédiaire d'une distance de wasserstein segmentée

Country Status (2)

Country Link
EP (1) EP3918527A1 (fr)
WO (1) WO2020159638A1 (fr)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113822228A (zh) * 2021-10-27 2021-12-21 南京大学 一种基于持续学习的用户表情识别方法和系统
CN114283287A (zh) * 2022-03-09 2022-04-05 南京航空航天大学 基于自训练噪声标签纠正的鲁棒领域自适应图像学习方法
EP4206740A1 (fr) 2021-12-30 2023-07-05 Yandex Self Driving Group Llc Procédé et système de détermination de degré de dégradation de données lidar

Non-Patent Citations (20)

* Cited by examiner, † Cited by third party
Title
ARJOVSKY, M.CHINTALA, S.BOTTOU, L., WASSERSTEIN GAN, 2017
COURTY, N.FLAMARY, R.TUIA, D.RAKOTOMAMONJY, A.: "Optimal transport for domain adaptation", IEEE TPAMI, vol. 39, no. 9, 2017, pages 1853 - 1865
GABOURIE, A.ROSTAMI, M.POPE, P.KOLOURI, S.KIM, K.: "Learning a domain-invariant embedding for unsupervised domain adaptation using class-conditioned distribution alignment", 2019 57TH ANNUAL ALLERTON CONFERENCE ON COMMUNICATION, CONTROL, AND COMPUTING, 2019
GANIN, Y.LEMPITSKY, V.: "Unsupervised domain adaptation by backpropagation", ICML, 2014
GHIFARY, M.KLEIJN, W. B.ZHANG, M.BALDUZZI, D.LI, W.: "Deep reconstruction-classification networks for unsupervised domain adaptation", EUROPEAN CONFERENCE ON COMPUTER VISION, 2016, pages 597 - 613, XP047355422, DOI: 10.1007/978-3-319-46493-0_36
HAGAN, M.T.MENHAJ, M.B.: "Training feedforward networks with the Marquardt algorithm", IEEE TRANSACTIONS ON NEURAL NETWORKS, vol. 5, no. 6, 1994, pages 989 - 993, XP000476812, DOI: 10.1109/72.329697
HOFFMAN, J.TZENG, E.PARK, T.ZHU, J.ISOLA, P.SAENKO, K.EFROS, A. A.DARRELL, T.: "Cycada: Cycle-consistent adversarial domain adaptation", ICML, 2018
JAEYOON YOO ET AL: "Domain Adaptation Using Adversarial Learning for Autonomous Navigation", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 11 December 2017 (2017-12-11), XP081320179 *
KOLOURI, S.MARTIN, C.E.ROHDE, G.K., SLICED-WASSERSTEIN AUTOENCODER: AN EMBARRASSINGLY SIMPLE GENERATIVE MODEL, 2018
KOLOURI, S.ROHDE, G.K.HOFFMAN, H.: "Sliced Wasserstein Distance for Learning Gaussian Mixture Models", CVPR, 2017
LIU, M.TUZEL, O.: "Coupled generative adversarial networks", IN ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS, 2016, pages 469 - 477
LUO, Z.ZOU, Y.HOFFMAN, J.FEI-FEI, L.F.: "Label Efficient Learning of Transferable Representations Across Domains and Tasks", ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS, 2017, pages 164 - 176
MOTIIAN SJONES QIRANMANESH SDORETTO G.: "Few-Shot Adversarial Domain Adaptation", ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS, 2017, pages 6673 - 6683
MUREZ, Z.KOLOURI, S.KRIEGMAN, D.RAMAMOORTHI, R.KIM, K.: "Image to Image Translation for Domain Adaptation", CVPR, 2017
PAN, S.J.YANG, Q.: "A survey on transfer learning", IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, vol. 22, no. 10, 2010, pages 1345 - 1359, XP055499519, DOI: 10.1109/TKDE.2009.191
SAITO, K.USHIKU, Y.HARADA, T.: "Asymmetric tri-training for unsupervised domain adaptation", ICML, 2018
SANKARANARAYANAN, S.BALAJI, Y.CASTILLO, C. D.CHELLAPPA, R.: "Generate to adapt: Aligning domains using generative adversarial networks", CVPR, 2018
SEGUY, V.DAMODARAN, B. B.FLAMARY, R.COURTY, N.ROLET, A.BLONDEL, M.: "Large-scale optimal transport and mapping estimation", ICLR, 2018
TZENG, E.HOFFMAN, J.SAENKO, K.DARRELL, T.: "Adversarial discriminative domain adaptation", COMPUTER VISION AND PATTERN RECOGNITION (CVPR, vol. 1, 2017, pages 4
ZOU YANG ET AL: "Unsupervised Domain Adaptation for Semantic Segmentation via Class-Balanced Self-training", 7 October 2018, ROBOCUP 2008: ROBOCUP 2008: ROBOT SOCCER WORLD CUP XII; [LECTURE NOTES IN COMPUTER SCIENCE; LECT.NOTES COMPUTER], SPRINGER INTERNATIONAL PUBLISHING, CHAM, PAGE(S) 297 - 313, ISBN: 978-3-319-10403-4, XP047498238 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113822228A (zh) * 2021-10-27 2021-12-21 南京大学 一种基于持续学习的用户表情识别方法和系统
CN113822228B (zh) * 2021-10-27 2024-03-22 南京大学 一种基于持续学习的用户表情识别方法和系统
EP4206740A1 (fr) 2021-12-30 2023-07-05 Yandex Self Driving Group Llc Procédé et système de détermination de degré de dégradation de données lidar
CN114283287A (zh) * 2022-03-09 2022-04-05 南京航空航天大学 基于自训练噪声标签纠正的鲁棒领域自适应图像学习方法

Also Published As

Publication number Publication date
EP3918527A1 (fr) 2021-12-08
CN113316790A (zh) 2021-08-27

Similar Documents

Publication Publication Date Title
US11176477B2 (en) System and method for unsupervised domain adaptation via sliced-wasserstein distance
Wicker et al. Feature-guided black-box safety testing of deep neural networks
Demirel et al. Zero-shot object detection by hybrid region embedding
Camara et al. Pedestrian models for autonomous driving Part I: low-level models, from sensing to tracking
US11288498B2 (en) Learning actions with few labels in the embedded space
US11023789B2 (en) Machine-vision method to classify input data based on object components
CN110073367B (zh) 用于零样本学习的利用基于soft-max的兼容性函数的多视图嵌入
CN113168567A (zh) 用于小样本转移学习的系统和方法
Sun et al. Application of BW-ELM model on traffic sign recognition
EP3918527A1 (fr) Système et procédé d'adaptation de domaine non supervisé par l'intermédiaire d'une distance de wasserstein segmentée
WO2019018022A2 (fr) Système de vision de machine sans donnée de référence par l'intermédiaire de représentations éparses communes
US11625557B2 (en) Process to learn new image classes without labels
US10691972B2 (en) Machine-vision system for discriminant localization of objects
WO2021118697A1 (fr) Procédé d'apprentissage de nouvelles classes d'images sans marqueurs
US10937150B2 (en) Systems and methods of feature correspondence analysis
WO2021133458A1 (fr) Systèmes et procédés d'apprentissage continu et sans supervision
Lai et al. Improving classification with semi-supervised and fine-grained learning
Khellal et al. Pedestrian classification and detection in far infrared images
Sahoo et al. RBI-2RCNN: Residual block intensity feature using a two-stage residual convolutional neural network for static hand gesture recognition
Pessanha Santos et al. Two‐stage 3D model‐based UAV pose estimation: A comparison of methods for optimization
Patel et al. Object detection with convolutional neural networks
Soleimani et al. Convolutional neural networks for aerial vehicle detection and recognition
Choi et al. Multi-UAV trajectory optimization and deep learning-based imagery analysis for a UAS-based inventory tracking solution
Wu et al. Small target recognition method on weak features
CN113316790B (zh) 用于自主学习代理中的无监督域适应的系统、方法和介质

Legal Events

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

Ref document number: 19842474

Country of ref document: EP

Kind code of ref document: A1

DPE1 Request for preliminary examination filed after expiration of 19th month from priority date (pct application filed from 20040101)
NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2019842474

Country of ref document: EP

Effective date: 20210830