Generative Hybrid Representations for Activity Forecasting with No-Regret Learning

被引:48
|
作者
Guan, Jiaqi [1 ,2 ]
Yuan, Ye [1 ]
Kitani, Kris M. [1 ]
Rhinehart, Nicholas [1 ,3 ]
机构
[1] Carnegie Mellon Univ, Pittsburgh, PA 15213 USA
[2] Univ Illinois, Champaign, IL 61820 USA
[3] Univ Calif Berkeley, Berkeley, CA USA
关键词
D O I
10.1109/CVPR42600.2020.00025
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Automatically reasoning about future human behaviors is a difficult problem but has significant practical applications to assistive systems. Part of this difficulty stems from learning systems' inability to represent all kinds of behaviors. Some behaviors, such as motion, are best described with continuous representations, whereas others, such as picking up a cup, are best described with discrete representations. Furthermore, human behavior is generally not fixed: people can change their habits and routines. This suggests these systems must be able to learn and adapt continuously. In this work, we develop an efficient deep generative model to jointly forecast a person's future discrete actions and continuous motions. On a large-scale egocentric dataset, EPIC-KITCHENS, we observe our method generates high-quality and diverse samples while exhibiting better generalization than related generative models. Finally, we propose a variant to continually learn our model from streaming data, observe its practical effectiveness, and theoretically justify its learning efficiency.
引用
收藏
页码:170 / 179
页数:10
相关论文
共 50 条
  • [1] Constrained no-regret learning
    Du, Ye
    Lehrer, Ehud
    JOURNAL OF MATHEMATICAL ECONOMICS, 2020, 88 : 16 - 24
  • [2] No-regret Reinforcement Learning
    Gopalan, Aditya
    2019 FIFTH INDIAN CONTROL CONFERENCE (ICC), 2019, : 16 - 16
  • [3] No-Regret Learning in Bayesian Games
    Hartline, Jason
    Syrgkanis, Vasilis
    Tardos, Eva
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 28 (NIPS 2015), 2015, 28
  • [4] No-Regret Learning Supports Voters' Competence
    Spelda, Petr
    Stritecky, Vit
    Symons, John
    SOCIAL EPISTEMOLOGY, 2024, 38 (05) : 543 - 559
  • [5] Limits and limitations of no-regret learning in games
    Monnot, Barnabe
    Piliouras, Georgios
    KNOWLEDGE ENGINEERING REVIEW, 2017, 32
  • [6] On the convergence of no-regret learning in selfish routing
    Krichene, Walid
    Drighes, Benjamin
    Bayen, Alexandre
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 32 (CYCLE 2), 2014, 32 : 163 - 171
  • [7] No-Regret Learning in Dynamic Stackelberg Games
    Lauffer, Niklas
    Ghasemi, Mahsa
    Hashemi, Abolfazl
    Savas, Yagiz
    Topcu, Ufuk
    IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2024, 69 (03) : 1418 - 1431
  • [8] Unifying convergence and no-regret in multiagent learning
    Banerjee, Bikramjit
    Peng, Jing
    LEARNING AND ADAPTION IN MULTI-AGENT SYSTEMS, 2006, 3898 : 100 - 114
  • [9] Weighted Voting Via No-Regret Learning
    Haghtalab, Nika
    Noothigattu, Ritesh
    Procaccia, Ariel D.
    THIRTY-SECOND AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE / THIRTIETH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE / EIGHTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2018, : 1055 - 1062
  • [10] No-regret Exploration in Contextual Reinforcement Learning
    Modi, Aditya
    Tewari, Ambuj
    CONFERENCE ON UNCERTAINTY IN ARTIFICIAL INTELLIGENCE (UAI 2020), 2020, 124 : 829 - 838