Hierarchies of Reward Machines

Daniel Furelos-Blanco, Mark Law, Anders Jonsson, Krysia Broda, Alessandra Russo
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:10494-10541, 2023.

Abstract

Reward machines (RMs) are a recent formalism for representing the reward function of a reinforcement learning task through a finite-state machine whose edges encode subgoals of the task using high-level events. The structure of RMs enables the decomposition of a task into simpler and independently solvable subtasks that help tackle long-horizon and/or sparse reward tasks. We propose a formalism for further abstracting the subtask structure by endowing an RM with the ability to call other RMs, thus composing a hierarchy of RMs (HRM). We exploit HRMs by treating each call to an RM as an independently solvable subtask using the options framework, and describe a curriculum-based method to learn HRMs from traces observed by the agent. Our experiments reveal that exploiting a handcrafted HRM leads to faster convergence than with a flat HRM, and that learning an HRM is feasible in cases where its equivalent flat representation is not.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-furelos-blanco23a, title = {Hierarchies of Reward Machines}, author = {Furelos-Blanco, Daniel and Law, Mark and Jonsson, Anders and Broda, Krysia and Russo, Alessandra}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {10494--10541}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/furelos-blanco23a/furelos-blanco23a.pdf}, url = {https://proceedings.mlr.press/v202/furelos-blanco23a.html}, abstract = {Reward machines (RMs) are a recent formalism for representing the reward function of a reinforcement learning task through a finite-state machine whose edges encode subgoals of the task using high-level events. The structure of RMs enables the decomposition of a task into simpler and independently solvable subtasks that help tackle long-horizon and/or sparse reward tasks. We propose a formalism for further abstracting the subtask structure by endowing an RM with the ability to call other RMs, thus composing a hierarchy of RMs (HRM). We exploit HRMs by treating each call to an RM as an independently solvable subtask using the options framework, and describe a curriculum-based method to learn HRMs from traces observed by the agent. Our experiments reveal that exploiting a handcrafted HRM leads to faster convergence than with a flat HRM, and that learning an HRM is feasible in cases where its equivalent flat representation is not.} }
Endnote
%0 Conference Paper %T Hierarchies of Reward Machines %A Daniel Furelos-Blanco %A Mark Law %A Anders Jonsson %A Krysia Broda %A Alessandra Russo %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-furelos-blanco23a %I PMLR %P 10494--10541 %U https://proceedings.mlr.press/v202/furelos-blanco23a.html %V 202 %X Reward machines (RMs) are a recent formalism for representing the reward function of a reinforcement learning task through a finite-state machine whose edges encode subgoals of the task using high-level events. The structure of RMs enables the decomposition of a task into simpler and independently solvable subtasks that help tackle long-horizon and/or sparse reward tasks. We propose a formalism for further abstracting the subtask structure by endowing an RM with the ability to call other RMs, thus composing a hierarchy of RMs (HRM). We exploit HRMs by treating each call to an RM as an independently solvable subtask using the options framework, and describe a curriculum-based method to learn HRMs from traces observed by the agent. Our experiments reveal that exploiting a handcrafted HRM leads to faster convergence than with a flat HRM, and that learning an HRM is feasible in cases where its equivalent flat representation is not.
APA
Furelos-Blanco, D., Law, M., Jonsson, A., Broda, K. & Russo, A.. (2023). Hierarchies of Reward Machines. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:10494-10541 Available from https://proceedings.mlr.press/v202/furelos-blanco23a.html.

Related Material