论文地址:http://proceedings.mlr.press/v37/schulman15.pdf
推荐几篇关于论文解读博客:
英文:
https://blog.csdn.net/xyp99/article/details/109378848
https://spinningup.openai.com/en/latest/algorithms/trpo.html
中文:
https://blog.csdn.net/qq_28385535/article/details/104892071
https://blog.csdn.net/kongcdy/article/details/102463598 (重要公式推导)
https://www.jianshu.com/p/34c2d8b31801
1. Introduction
主要写了三个方面内容,一是对策略优化方法的分类、二是三种方法各自的优缺点、三是对论文思路的概括</