The Sample-Complexity of General Reinforcement Learning

Tor Lattimore, Marcus Hutter, Peter Sunehag
; Proceedings of the 30th International Conference on Machine Learning, PMLR 28(3):28-36, 2013.

Abstract

We study the sample-complexity of reinforcement learning in a general setting without assuming ergodicity or finiteness of the environment. Instead, we define a topology on the space of environments and show that if an environment class is compact with respect to this topology then finite sample-complexity bounds are possible and give an algorithm achieving these bounds. We also show the existence of environment classes that are non-compact where finite sample-complexity bounds are not achievable. A lower bound is presented that matches the upper bound except for logarithmic factors.

Cite this Paper


BibTeX
@InProceedings{pmlr-v28-lattimore13, title = {The Sample-Complexity of General Reinforcement Learning}, author = {Tor Lattimore and Marcus Hutter and Peter Sunehag}, booktitle = {Proceedings of the 30th International Conference on Machine Learning}, pages = {28--36}, year = {2013}, editor = {Sanjoy Dasgupta and David McAllester}, volume = {28}, number = {3}, series = {Proceedings of Machine Learning Research}, address = {Atlanta, Georgia, USA}, month = {17--19 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v28/lattimore13.pdf}, url = {http://proceedings.mlr.press/v28/lattimore13.html}, abstract = {We study the sample-complexity of reinforcement learning in a general setting without assuming ergodicity or finiteness of the environment. Instead, we define a topology on the space of environments and show that if an environment class is compact with respect to this topology then finite sample-complexity bounds are possible and give an algorithm achieving these bounds. We also show the existence of environment classes that are non-compact where finite sample-complexity bounds are not achievable. A lower bound is presented that matches the upper bound except for logarithmic factors. } }
Endnote
%0 Conference Paper %T The Sample-Complexity of General Reinforcement Learning %A Tor Lattimore %A Marcus Hutter %A Peter Sunehag %B Proceedings of the 30th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2013 %E Sanjoy Dasgupta %E David McAllester %F pmlr-v28-lattimore13 %I PMLR %J Proceedings of Machine Learning Research %P 28--36 %U http://proceedings.mlr.press %V 28 %N 3 %W PMLR %X We study the sample-complexity of reinforcement learning in a general setting without assuming ergodicity or finiteness of the environment. Instead, we define a topology on the space of environments and show that if an environment class is compact with respect to this topology then finite sample-complexity bounds are possible and give an algorithm achieving these bounds. We also show the existence of environment classes that are non-compact where finite sample-complexity bounds are not achievable. A lower bound is presented that matches the upper bound except for logarithmic factors.
RIS
TY - CPAPER TI - The Sample-Complexity of General Reinforcement Learning AU - Tor Lattimore AU - Marcus Hutter AU - Peter Sunehag BT - Proceedings of the 30th International Conference on Machine Learning PY - 2013/02/13 DA - 2013/02/13 ED - Sanjoy Dasgupta ED - David McAllester ID - pmlr-v28-lattimore13 PB - PMLR SP - 28 DP - PMLR EP - 36 L1 - http://proceedings.mlr.press/v28/lattimore13.pdf UR - http://proceedings.mlr.press/v28/lattimore13.html AB - We study the sample-complexity of reinforcement learning in a general setting without assuming ergodicity or finiteness of the environment. Instead, we define a topology on the space of environments and show that if an environment class is compact with respect to this topology then finite sample-complexity bounds are possible and give an algorithm achieving these bounds. We also show the existence of environment classes that are non-compact where finite sample-complexity bounds are not achievable. A lower bound is presented that matches the upper bound except for logarithmic factors. ER -
APA
Lattimore, T., Hutter, M. & Sunehag, P.. (2013). The Sample-Complexity of General Reinforcement Learning. Proceedings of the 30th International Conference on Machine Learning, in PMLR 28(3):28-36

Related Material