arrow
返回

Guided deterministic policy optimization with gradient-free policy parameters information

delete2023-11-01
delete2
PRE
AI
沈淳 (Chun Shen)
S
Sheng Zhu
S
Shuai D. Han
X
Xiaoyu Gong
S
Shuai Lü *
DOI:10.1016/j.eswa.2023.120693delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Deep Deterministic Policy Gradient (DDPG) and Twin Delayed Deep Deterministic Policy Gradient (TD3) are two classical deterministic policy gradient algorithms. It is worth noting that the policies of both DDPG and TD3 are completely dependent on the gradient of critics. This will cause the policy to be unstable and easy to converge to the local optimum in the learning process. Although the idea of maximum entropy learning can provide more effective exploration, it can only be applied to the algorithm using stochastic policy, not to DDPG and TD3. In this paper, we propose a deterministic policy optimization method combining gradient-free policy parameters information (GFPPI). Specifically, we obtain a new set of policies by injecting Gaussian noise into the policy parameters, and then weight these policy parameters based on critics to obtain GFPPI. Finally, GFPPI is used as the regularization term of the policy optimization function to guide the policy update. GFPPI can mitigate premature policy convergence and facilitate exploration with optimistic principles. We provide the theoretical guarantee for monotonic improvement of expected cumulative return using augmented loss function with GFPPI, experimentally analyze the role of GFPPI in policy optimization and combine it with deterministic policy gradient information for policy optimization. The experiments on OpenAI gym demonstrate that GFPPI can improve sample efficiency and enable the algorithm to get higher performance.
Keyword:
Deterministic policy gradient
Premature convergence
Local optimum
Policy optimization
Exploration
Sample efficiency

期刊

Expert Systems with Applications 封面图
Expert Systems with Applications
IF:
7.5
论文数:
2.9W
被引数:
10.2W

机构

J
Jilin University
学者数:
8.7W
论文数: 5.6W
被引数: 8.9K
引用论文

引用论文

The Economic Case for Protecting Biodiversity
err
IF0
err2020-10-01
err0
errOAAI
errGeoffrey Heal
err分享
err收藏
Dynamics of polymer film formation during spin coating
err2014-09-26
err0
errOAAI
errY. Mouhamad; P. Mokarian-Tabari; N. Clarke; R. A. L. Jones; M. Geoghegan
err分享
err收藏
Depression During Pregnancy
err2010-11-30
err0
PREAI
errPriscille Gerardin; Jaqueline Wendland; Nicolas Bodeau; Armelle Galin; Stéphanie Bialobos; Sylvie Tordjman; Philippe Mazet; Yves Darbois; Jacky Nizard; Marc Dommergues; David Cohen
err分享
err收藏
没有更多内容