Offline RL methods adapted to factorised action spaces match or beat behaviour cloning and are more scalable than atomic action representations.
Learning action representations for reinforcement learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
An Investigation of Offline Reinforcement Learning in Factorisable Action Spaces
Offline RL methods adapted to factorised action spaces match or beat behaviour cloning and are more scalable than atomic action representations.