Robust Imitation Learning from Noisy Demonstrations

Voot Tangkaratt, Nontawat Charoenphakdee, Masashi Sugiyama
Proceedings of The 24th International Conference on Artificial Intelligence and Statistics, PMLR 130:298-306, 2021.

Abstract

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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v130-tangkaratt21a, title = { Robust Imitation Learning from Noisy Demonstrations }, author = {Tangkaratt, Voot and Charoenphakdee, Nontawat and Sugiyama, Masashi}, booktitle = {Proceedings of The 24th International Conference on Artificial Intelligence and Statistics}, pages = {298--306}, year = {2021}, editor = {Banerjee, Arindam and Fukumizu, Kenji}, volume = {130}, series = {Proceedings of Machine Learning Research}, month = {13--15 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v130/tangkaratt21a/tangkaratt21a.pdf}, url = {https://proceedings.mlr.press/v130/tangkaratt21a.html}, abstract = { 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. } }
Endnote
%0 Conference Paper %T Robust Imitation Learning from Noisy Demonstrations %A Voot Tangkaratt %A Nontawat Charoenphakdee %A Masashi Sugiyama %B Proceedings of The 24th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2021 %E Arindam Banerjee %E Kenji Fukumizu %F pmlr-v130-tangkaratt21a %I PMLR %P 298--306 %U https://proceedings.mlr.press/v130/tangkaratt21a.html %V 130 %X 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.
APA
Tangkaratt, V., Charoenphakdee, N. & Sugiyama, M.. (2021). Robust Imitation Learning from Noisy Demonstrations . Proceedings of The 24th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 130:298-306 Available from https://proceedings.mlr.press/v130/tangkaratt21a.html.

Related Material