Robust Imitation Learning from Noisy Demonstrations

被引：0

作者：

Tangkaratt, Voot ^{[1
]}

Charoenphakdee, Nontawat ^{[1
,2
]}

Sugiyama, Masashi ^{[1
,2
]}

机构：

[1] RIKEN, Tokyo, Japan

[2] Univ Tokyo, Tokyo, Japan

来源：

24TH INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS (AISTATS) | 2021年 / 130卷

关键词：

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Robust learning from noisy demonstrations is a practical but highly challenging problem in imitation learning. In this paper, we first theoretically show that robust imitation learning can be achieved by optimizing a classification risk with a symmetric loss. Based on this theoretical finding, we then propose a new imitation learning method that optimizes the classification risk by effectively combining pseudo-labeling with co-training. Unlike existing methods, our method does not require additional labels or strict assumptions about noise distributions. Experimental results on continuous-control benchmarks show that our method is more robust compared to state-of-the-art methods.

引用

页码：298 / +

页数：11