Sample Efficient Actor-Critic with Experience ReplayDownload PDF

21 Oct 2020 (modified: 23 Feb 2017)ICLR 2017 conference submissionReaders: Everyone
  • TL;DR: Prepared for ICLR 2017.
  • Abstract: This paper presents an actor-critic deep reinforcement learning agent with experience replay that is stable, sample efficient, and performs remarkably well on challenging environments, including the discrete 57-game Atari domain and several continuous control problems. To achieve this, the paper introduces several innovations, including truncated importance sampling with bias correction, stochastic dueling network architectures, and a new trust region policy optimization method.
  • Keywords: Deep learning, Reinforcement Learning
  • Conflicts: google.com, ox.ac.uk, ubc.ca
19 Replies

Loading