IQL achieves policy improvement in offline RL by implicitly estimating optimal action values through state-conditional upper expectiles of value functions, without querying Q-functions on out-of-distribution actions.
Stabilizing off-policy q-learning via bootstrapping error reduction.arXiv preprint arXiv:1906.00949
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
representative citing papers
D4RL supplies new offline RL benchmarks and datasets from expert and mixed sources to expose weaknesses in existing algorithms and standardize evaluation.
RankQ adds a self-supervised ranking loss to Q-learning to learn structured action orderings, yielding competitive or better performance than prior methods on D4RL benchmarks and large gains in vision-based robot fine-tuning.
Behavior-regularized actor-critic methods achieve strong offline RL results with simple regularization, rendering many recent technical additions unnecessary.
AWR learns policies via advantage-weighted supervised regression on actions, achieving competitive off-policy performance on Gym tasks and strong results from static data alone.
citing papers explorer
-
Offline Reinforcement Learning with Implicit Q-Learning
IQL achieves policy improvement in offline RL by implicitly estimating optimal action values through state-conditional upper expectiles of value functions, without querying Q-functions on out-of-distribution actions.
-
D4RL: Datasets for Deep Data-Driven Reinforcement Learning
D4RL supplies new offline RL benchmarks and datasets from expert and mixed sources to expose weaknesses in existing algorithms and standardize evaluation.
-
RankQ: Offline-to-Online Reinforcement Learning via Self-Supervised Action Ranking
RankQ adds a self-supervised ranking loss to Q-learning to learn structured action orderings, yielding competitive or better performance than prior methods on D4RL benchmarks and large gains in vision-based robot fine-tuning.
-
Behavior Regularized Offline Reinforcement Learning
Behavior-regularized actor-critic methods achieve strong offline RL results with simple regularization, rendering many recent technical additions unnecessary.
-
Advantage-Weighted Regression: Simple and Scalable Off-Policy Reinforcement Learning
AWR learns policies via advantage-weighted supervised regression on actions, achieving competitive off-policy performance on Gym tasks and strong results from static data alone.