Dynamic Programming Boosting for Discriminative Macro-Action Discovery

Leonidas Lefakis, Francois Fleuret
Proceedings of the 31st International Conference on Machine Learning, PMLR 32(2):1548-1556, 2014.

Abstract

We consider the problem of automatic macro-action discovery in imitation learning, which we cast as one of change-point detection. Unlike prior work in change-point detection, the present work leverages discriminative learning algorithms. Our main contribution is a novel supervised learning algorithm which extends the classical Boosting framework by combining it with dynamic programming. The resulting process alternatively improves the performance of individual strong predictors and the estimated change-points in the training sequence. Empirical evaluation is presented for the proposed method on tasks where change-points arise naturally as part of a classification problem. Finally we show the applicability of the algorithm to macro-action discovery in imitation learning and demonstrate it allows us to solve complex image-based goal-planning problems with thousands of features.

Cite this Paper

BibTeX
@InProceedings{pmlr-v32-lefakis14, title = {Dynamic Programming Boosting for Discriminative Macro-Action Discovery}, author = {Lefakis, Leonidas and Fleuret, Francois}, booktitle = {Proceedings of the 31st International Conference on Machine Learning}, pages = {1548--1556}, year = {2014}, editor = {Xing, Eric P. and Jebara, Tony}, volume = {32}, number = {2}, series = {Proceedings of Machine Learning Research}, address = {Bejing, China}, month = {22--24 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v32/lefakis14.pdf}, url = {https://proceedings.mlr.press/v32/lefakis14.html}, abstract = {We consider the problem of automatic macro-action discovery in imitation learning, which we cast as one of change-point detection. Unlike prior work in change-point detection, the present work leverages discriminative learning algorithms. Our main contribution is a novel supervised learning algorithm which extends the classical Boosting framework by combining it with dynamic programming. The resulting process alternatively improves the performance of individual strong predictors and the estimated change-points in the training sequence. Empirical evaluation is presented for the proposed method on tasks where change-points arise naturally as part of a classification problem. Finally we show the applicability of the algorithm to macro-action discovery in imitation learning and demonstrate it allows us to solve complex image-based goal-planning problems with thousands of features.} }
Endnote
%0 Conference Paper %T Dynamic Programming Boosting for Discriminative Macro-Action Discovery %A Leonidas Lefakis %A Francois Fleuret %B Proceedings of the 31st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2014 %E Eric P. Xing %E Tony Jebara %F pmlr-v32-lefakis14 %I PMLR %P 1548--1556 %U https://proceedings.mlr.press/v32/lefakis14.html %V 32 %N 2 %X We consider the problem of automatic macro-action discovery in imitation learning, which we cast as one of change-point detection. Unlike prior work in change-point detection, the present work leverages discriminative learning algorithms. Our main contribution is a novel supervised learning algorithm which extends the classical Boosting framework by combining it with dynamic programming. The resulting process alternatively improves the performance of individual strong predictors and the estimated change-points in the training sequence. Empirical evaluation is presented for the proposed method on tasks where change-points arise naturally as part of a classification problem. Finally we show the applicability of the algorithm to macro-action discovery in imitation learning and demonstrate it allows us to solve complex image-based goal-planning problems with thousands of features.
RIS
TY - CPAPER TI - Dynamic Programming Boosting for Discriminative Macro-Action Discovery AU - Leonidas Lefakis AU - Francois Fleuret BT - Proceedings of the 31st International Conference on Machine Learning DA - 2014/06/18 ED - Eric P. Xing ED - Tony Jebara ID - pmlr-v32-lefakis14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 32 IS - 2 SP - 1548 EP - 1556 L1 - http://proceedings.mlr.press/v32/lefakis14.pdf UR - https://proceedings.mlr.press/v32/lefakis14.html AB - We consider the problem of automatic macro-action discovery in imitation learning, which we cast as one of change-point detection. Unlike prior work in change-point detection, the present work leverages discriminative learning algorithms. Our main contribution is a novel supervised learning algorithm which extends the classical Boosting framework by combining it with dynamic programming. The resulting process alternatively improves the performance of individual strong predictors and the estimated change-points in the training sequence. Empirical evaluation is presented for the proposed method on tasks where change-points arise naturally as part of a classification problem. Finally we show the applicability of the algorithm to macro-action discovery in imitation learning and demonstrate it allows us to solve complex image-based goal-planning problems with thousands of features. ER -
APA
Lefakis, L. & Fleuret, F.. (2014). Dynamic Programming Boosting for Discriminative Macro-Action Discovery. Proceedings of the 31st International Conference on Machine Learning, in Proceedings of Machine Learning Research 32(2):1548-1556 Available from https://proceedings.mlr.press/v32/lefakis14.html.

Related Material