Generalized Compatible Function Approximation for Policy Gradient Search

Yiming Peng, Gang Chen, Mengjie Zhang, Shaoning PangICONIP 2016 · OpenReview

Abstract

Reinforcement learning aims at solving stochastic sequential decision making problems through direct trial-and-error interactions with the learning environment. In this paper, we will develop generalized compatible features to approximate value functions for reliable Reinforcement Learning. Further guided by an Actor-Critic Reinforcement Learning paradigm, we will also develop a generalized updating rule for policy gradient search in order to constantly improve learning performance. Our new updating rule has been examined on several benchmark learning problems. The experimental results on two problems will be reported specifically in this paper. Our results show that, under suitable generalization of the updating rule, the learning performance and reliability can be noticeably improved.
BibTeX
@inproceedings{peng2016generalized,
  author = {Yiming Peng and Gang Chen and Mengjie Zhang and Shaoning Pang},
  title = {{Generalized Compatible Function Approximation for Policy Gradient Search}},
  booktitle = {ICONIP},
  year = {2016},
  url = {https://openreview.net/forum?id=LSq61HsuY6O},
}

Loading related content...

Figure