Detecting Hidden Triggers: Mapping Non-Markov Reward Functions to Markov

arXiv:2401.11325v2 Announce Type: replace
Abstract: Many Reinforcement Learning algorithms assume a Markov reward function to guarantee optimality. However, not all reward functions are known to be Markov. In this paper, we propose a framework for mapping non-Markov reward functions into equivalent Markov ones by learning a Reward Machine - a specialized reward automaton. Unlike the general practice of learning Reward Machines, we do not require a set of high-level propositional symbols from which to learn. Rather, we learn \emph{hidden triggers} …

abstract algorithms arxiv cs.ai cs.lg framework function functions hidden however machine mapping markov ones paper reinforcement reinforcement learning type