Reinforcement Learning with General Value Function Approximation: Provably Efficient Approach via Bounded Eluder Dimension

被引:0
作者
Wang, Ruosong [1 ]
Salakhutdinov, Ruslan [1 ]
Yang, Lin F. [2 ]
机构
[1] Carnegie Mellon Univ, Pittsburgh, PA 15213 USA
[2] Univ Calif Los Angeles, Los Angeles, CA USA
来源
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 33, NEURIPS 2020 | 2020年 / 33卷
关键词
PAC BOUNDS;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Value function approximation has demonstrated phenomenal empirical success in reinforcement learning (RL). Nevertheless, despite a handful of recent progress on developing theory for RL with linear function approximation, the understanding of general function approximation schemes largely remains missing. In this paper, we establish the first provably efficient RL algorithm with general value function approximation. We show that if the value functions admit an approximation with a function class F, our algorithm achieves a regret bound of (O) over tilde (poly(dH)root T) where d is a complexity measure of F that depends on the eluder dimension [Russo and Van Roy, 2013] and log-covering numbers, H is the planning horizon, and T is the number interactions with the environment. Our theory generalizes the linear MDP assumption to general function classes. Moreover, our algorithm is model-free and provides a framework to justify the effectiveness of algorithms used in practice.
引用
收藏
页数:13
相关论文
共 69 条
  • [51] Sidford Aaron, 2018, ARXIV180601492
  • [52] Mastering the game of Go without human knowledge
    Silver, David
    Schrittwieser, Julian
    Simonyan, Karen
    Antonoglou, Ioannis
    Huang, Aja
    Guez, Arthur
    Hubert, Thomas
    Baker, Lucas
    Lai, Matthew
    Bolton, Adrian
    Chen, Yutian
    Lillicrap, Timothy
    Hui, Fan
    Sifre, Laurent
    van den Driessche, George
    Graepel, Thore
    Hassabis, Demis
    [J]. NATURE, 2017, 550 (7676) : 354 - +
  • [53] Singh D, 2020, AGROCHEMICALS DETECTION, TREATMENT AND REMEDIATION: PESTICIDES AND CHEMICAL FERTILIZERS, P101, DOI 10.1016/B978-0-08-103017-2.00004-0
  • [54] GRAPH SPARSIFICATION BY EFFECTIVE RESISTANCES
    Spielman, Daniel A.
    Srivastava, Nikhil
    [J]. SIAM JOURNAL ON COMPUTING, 2011, 40 (06) : 1913 - 1926
  • [55] Strehl A. L., 2006, P 23 INT C MACH LEAR, P881, DOI DOI 10.1242/jcs.036517
  • [56] Strehl AL, 2009, J MACH LEARN RES, V10, P2413
  • [57] Sun W., 2019, PMLR, P2898
  • [58] Szepesvari Csaba, 2005, P 22 INT C MACH LEAR, P880, DOI [DOI 10.1145/1102351.1102462, 10.1145/1102351.1102462]
  • [59] Szita I., 2010, ICML, P1031
  • [60] van Hasselt H, 2016, AAAI CONF ARTIF INTE, P2094