arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1802.07564cs.LGcs.AIstat.ML
收藏

裁剪操作策略梯度

Clipped Action Policy Gradient

Preferred Networks, Inc.(Preferred Networks公司)

Yasuhiro Fujita, Shin-ichi Maeda

摘要:

许多连续控制任务具有有界的动作空间。当将策略梯度方法应用于此类任务时,需要在执行前裁剪越界操作,而策略通常会被优化,就像操作未被裁剪一样。我们提出了一个策略梯度估计,利用知识的行动被裁剪,以减少估计的方差。我们证明了我们的估计,命名为剪切动作策略梯度(CAPG),是无偏的,并实现较低的方差比传统的估计,忽略行动的界限。实验结果表明,CAPG一般优于传统的估计,这表明它是一个更好的连续控制任务的策略梯度估计。源代码可在https://github.com/pfnet-research/capg上获得。

英文摘要:

Many continuous control tasks have bounded action spaces. When policy gradient methods are applied to such tasks, out-of-bound actions need to be clipped before execution, while policies are usually optimized as if the actions are not clipped. We propose a policy gradient estimator that exploits the knowledge of actions being clipped to reduce the variance in estimation. We prove that our estimator, named clipped action policy gradient (CAPG), is unbiased and achieves lower variance than the conventional estimator that ignores action bounds. Experimental results demonstrate that CAPG generally outperforms the conventional estimator, indicating that it is a better policy gradient estimator for continuous control tasks. The source code is available at https://github.com/pfnet-research/capg.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑