Reward machines: Exploiting reward function structure in reinforcement learning

RT Icarte, TQ Klassen, R Valenzano… - Journal of Artificial …, 2022 - jair.org
Reinforcement learning (RL) methods usually treat reward functions as black boxes. As
such, these methods must extensively interact with the environment in order to discover …

Learning reward machines for partially observable reinforcement learning

R Toro Icarte, E Waldie, T Klassen… - Advances in neural …, 2019 - proceedings.neurips.cc
Abstract Reward Machines (RMs), originally proposed for specifying problems in
Reinforcement Learning (RL), provide a structured, automata-based representation of a …

Interpretable sequence classification via discrete optimization

M Shvo, AC Li, RT Icarte, SA McIlraith - Proceedings of the AAAI …, 2021 - ojs.aaai.org
Sequence classification is the task of predicting a class label given a sequence of
observations. In many applications such as healthcare monitoring or intrusion detection …

Learning non-Markovian reward models in MDPs

G Rens, JF Raskin - arxiv preprint arxiv:2001.09293, 2020 - arxiv.org
There are situations in which an agent should receive rewards only after having
accomplished a series of previous tasks. In other words, the reward that the agent receives …

Reward Machines

RAT Icarte - 2022 - search.proquest.com
Reinforcement learning involves the study of how to solve sequential decision-making
problems using minimal supervision or prior knowledge. In contrast to most methods for …

Reward Machines

RA Toro Icarte - 2022 - tspace.library.utoronto.ca
Reinforcement learning involves the study of how to solve sequential decision-making
problems using minimal supervision or prior knowledge. In contrast to most methods for …