Unifying Task Specification in Reinforcement Learning

Martha White
Proceedings of the 34th International Conference on Machine Learning, PMLR 70:3742-3750, 2017.

Abstract

Reinforcement learning tasks are typically specified as Markov decision processes. This formalism has been highly successful, though specifications often couple the dynamics of the environment and the learning objective. This lack of modularity can complicate generalization of the task specification, as well as obfuscate connections between different task settings, such as episodic and continuing. In this work, we introduce the RL task formalism, that provides a unification through simple constructs including a generalization to transition-based discounting. Through a series of examples, we demonstrate the generality and utility of this formalism. Finally, we extend standard learning constructs, including Bellman operators, and extend some seminal theoretical results, including approximation errors bounds. Overall, we provide a well-understood and sound formalism on which to build theoretical results and simplify algorithm use and development.

Cite this Paper


BibTeX
@InProceedings{pmlr-v70-white17a, title = {Unifying Task Specification in Reinforcement Learning}, author = {Martha White}, booktitle = {Proceedings of the 34th International Conference on Machine Learning}, pages = {3742--3750}, year = {2017}, editor = {Precup, Doina and Teh, Yee Whye}, volume = {70}, series = {Proceedings of Machine Learning Research}, month = {06--11 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v70/white17a/white17a.pdf}, url = { http://proceedings.mlr.press/v70/white17a.html }, abstract = {Reinforcement learning tasks are typically specified as Markov decision processes. This formalism has been highly successful, though specifications often couple the dynamics of the environment and the learning objective. This lack of modularity can complicate generalization of the task specification, as well as obfuscate connections between different task settings, such as episodic and continuing. In this work, we introduce the RL task formalism, that provides a unification through simple constructs including a generalization to transition-based discounting. Through a series of examples, we demonstrate the generality and utility of this formalism. Finally, we extend standard learning constructs, including Bellman operators, and extend some seminal theoretical results, including approximation errors bounds. Overall, we provide a well-understood and sound formalism on which to build theoretical results and simplify algorithm use and development.} }
Endnote
%0 Conference Paper %T Unifying Task Specification in Reinforcement Learning %A Martha White %B Proceedings of the 34th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2017 %E Doina Precup %E Yee Whye Teh %F pmlr-v70-white17a %I PMLR %P 3742--3750 %U http://proceedings.mlr.press/v70/white17a.html %V 70 %X Reinforcement learning tasks are typically specified as Markov decision processes. This formalism has been highly successful, though specifications often couple the dynamics of the environment and the learning objective. This lack of modularity can complicate generalization of the task specification, as well as obfuscate connections between different task settings, such as episodic and continuing. In this work, we introduce the RL task formalism, that provides a unification through simple constructs including a generalization to transition-based discounting. Through a series of examples, we demonstrate the generality and utility of this formalism. Finally, we extend standard learning constructs, including Bellman operators, and extend some seminal theoretical results, including approximation errors bounds. Overall, we provide a well-understood and sound formalism on which to build theoretical results and simplify algorithm use and development.
APA
White, M.. (2017). Unifying Task Specification in Reinforcement Learning. Proceedings of the 34th International Conference on Machine Learning, in Proceedings of Machine Learning Research 70:3742-3750 Available from http://proceedings.mlr.press/v70/white17a.html .

Related Material