Exact Reduction of Huge Action Spaces in General Reinforcement Learning

Authors

  • Sultan J. Majeed Australian National University
  • Marcus Hutter DeepMind Australian National University

Keywords:

Reinforcement Learning

Abstract

The reinforcement learning (RL) framework formalizes the notion of learning with interactions. Many real-world problems have large state-spaces and/or action-spaces such as in Go, StarCraft, protein folding, and robotics or are non-Markovian, which cause significant challenges to RL algorithms. In this work we address the large action-space problem by sequentializing actions, which can reduce the action-space size significantly, even down to two actions at the expense of an increased planning horizon. We provide explicit and exact constructions and equivalence proofs for all quantities of interest for arbitrary history-based processes. In the case of MDPs, this could help RL algorithms that bootstrap. In this work we show how action-binarization in the non-MDP case can significantly improve Extreme State Aggregation (ESA) bounds. ESA allows casting any (non-MDP, non-ergodic, history-based) RL problem into a fixed-sized non-Markovian state-space with the help of a surrogate Markovian process. On the upside, ESA enjoys similar optimality guarantees as Markovian models do. But a downside is that the size of the aggregated state-space becomes exponential in the size of the action-space. In this work, we patch this issue by binarizing the action-space. We provide an upper bound on the number of states of this binarized ESA that is logarithmic in the original action-space size, a double-exponential improvement.

Downloads

Published

2021-05-18

How to Cite

Majeed, S. J., & Hutter, M. (2021). Exact Reduction of Huge Action Spaces in General Reinforcement Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 35(10), 8874-8883. Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/17074

Issue

Section

AAAI Technical Track on Machine Learning III