Autonomous Exploration For Navigating In MDPs

Shiau Hong Lim, Peter Auer
Proceedings of the 25th Annual Conference on Learning Theory, PMLR 23:40.1-40.24, 2012.

Abstract

While intrinsically motivated learning agents hold considerable promise to overcome limitations of more supervised learning systems, quantitative evaluation and theoretical analysis of such agents are difficult. We propose to consider a restricted setting for autonomous learning where systematic evaluation of learning performance is possible. In this setting the agent needs to learn to navigate in a Markov Decision Process where extrinsic rewards are not present or are ignored. We present a learning algorithm for this scenario and evaluate it by the amount of exploration it uses to learn the environment.

Cite this Paper

BibTeX
@InProceedings{pmlr-v23-lim12, title = {Autonomous Exploration For Navigating In MDPs}, author = {Lim, Shiau Hong and Auer, Peter}, booktitle = {Proceedings of the 25th Annual Conference on Learning Theory}, pages = {40.1--40.24}, year = {2012}, editor = {Mannor, Shie and Srebro, Nathan and Williamson, Robert C.}, volume = {23}, series = {Proceedings of Machine Learning Research}, address = {Edinburgh, Scotland}, month = {25--27 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v23/lim12/lim12.pdf}, url = {https://proceedings.mlr.press/v23/lim12.html}, abstract = {While intrinsically motivated learning agents hold considerable promise to overcome limitations of more supervised learning systems, quantitative evaluation and theoretical analysis of such agents are difficult. We propose to consider a restricted setting for autonomous learning where systematic evaluation of learning performance is possible. In this setting the agent needs to learn to navigate in a Markov Decision Process where extrinsic rewards are not present or are ignored. We present a learning algorithm for this scenario and evaluate it by the amount of exploration it uses to learn the environment.} }
Endnote
%0 Conference Paper %T Autonomous Exploration For Navigating In MDPs %A Shiau Hong Lim %A Peter Auer %B Proceedings of the 25th Annual Conference on Learning Theory %C Proceedings of Machine Learning Research %D 2012 %E Shie Mannor %E Nathan Srebro %E Robert C. Williamson %F pmlr-v23-lim12 %I PMLR %P 40.1--40.24 %U https://proceedings.mlr.press/v23/lim12.html %V 23 %X While intrinsically motivated learning agents hold considerable promise to overcome limitations of more supervised learning systems, quantitative evaluation and theoretical analysis of such agents are difficult. We propose to consider a restricted setting for autonomous learning where systematic evaluation of learning performance is possible. In this setting the agent needs to learn to navigate in a Markov Decision Process where extrinsic rewards are not present or are ignored. We present a learning algorithm for this scenario and evaluate it by the amount of exploration it uses to learn the environment.
RIS
TY - CPAPER TI - Autonomous Exploration For Navigating In MDPs AU - Shiau Hong Lim AU - Peter Auer BT - Proceedings of the 25th Annual Conference on Learning Theory DA - 2012/06/16 ED - Shie Mannor ED - Nathan Srebro ED - Robert C. Williamson ID - pmlr-v23-lim12 PB - PMLR DP - Proceedings of Machine Learning Research VL - 23 SP - 40.1 EP - 40.24 L1 - http://proceedings.mlr.press/v23/lim12/lim12.pdf UR - https://proceedings.mlr.press/v23/lim12.html AB - While intrinsically motivated learning agents hold considerable promise to overcome limitations of more supervised learning systems, quantitative evaluation and theoretical analysis of such agents are difficult. We propose to consider a restricted setting for autonomous learning where systematic evaluation of learning performance is possible. In this setting the agent needs to learn to navigate in a Markov Decision Process where extrinsic rewards are not present or are ignored. We present a learning algorithm for this scenario and evaluate it by the amount of exploration it uses to learn the environment. ER -
APA
Lim, S.H. & Auer, P.. (2012). Autonomous Exploration For Navigating In MDPs. Proceedings of the 25th Annual Conference on Learning Theory, in Proceedings of Machine Learning Research 23:40.1-40.24 Available from https://proceedings.mlr.press/v23/lim12.html.

Related Material