方策対照的模倣学習
Policy Contrastive Imitation Learning
敵対的模倣学習の識別器表現の質が低い問題を解決するため、異なる方策を基準にした対照的表現空間を学習し、コサイン類似度に基づく滑らかな報酬を生成するPCILを提案した。
著者: Jialei Huang, Zhaoheng Yin, Yingdong Hu, Yang Gao
分類: cs.LG, cs.RO
原文アブストラクト
Adversarial imitation learning (AIL) is a popular method that has recently achieved much success. However, the performance of AIL is still unsatisfactory on the more challenging tasks. We find that one of the major reasons is due to the low quality of AIL discriminator representation. Since the AIL discriminator is trained via binary classification that does not necessarily discriminate the policy from the expert in a meaningful way, the resulting reward might not be meaningful either. We propose a new method called Policy Contrastive Imitation Learning (PCIL) to resolve this issue. PCIL learns a contrastive representation space by anchoring on different policies and generates a smooth cosine-similarity-based reward. Our proposed representation learning objective can be viewed as a stronger version of the AIL objective and provide a more meaningful comparison between the agent and the policy. From a theoretical perspective, we show the validity of our method using the apprenticeship learning framework. Furthermore, our empirical evaluation on the DeepMind Control suite demonstrates that PCIL can achieve state-of-the-art performance. Finally, qualitative results suggest that PCIL builds a smoother and more meaningful representation space for imitation learning.
