A two-layer MLP with recursive skip-step sub-goal planning can reach state-of-the-art offline RL scores on long-horizon D4RL tasks, challenging the need for large expressive models.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Are Expressive Models Truly Necessary for Offline RL?
A two-layer MLP with recursive skip-step sub-goal planning can reach state-of-the-art offline RL scores on long-horizon D4RL tasks, challenging the need for large expressive models.