Q-DOT uses gradient maps of input-convex neural networks to compute Wasserstein regularization in offline RL, achieving D4RL scores comparable to or better than IQL without adversarial training.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Offline Reinforcement Learning with Wasserstein Regularization via Optimal Transport Maps
Q-DOT uses gradient maps of input-convex neural networks to compute Wasserstein regularization in offline RL, achieving D4RL scores comparable to or better than IQL without adversarial training.