EXPO stabilizes online RL for expressive policies by training a base policy with imitation and using a lightweight Gaussian edit policy to select higher-value actions on the fly for sampling and TD backups.
Cal-ql: Calibrated offline rl pre-training for efficient online fine-tuning
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 3roles
background 1polarities
support 1representative citing papers
OGPO is a sample-efficient off-policy method for full finetuning of generative control policies that reaches SOTA on robotic manipulation tasks and can recover from poor behavior-cloning initializations without expert data.
IDQL generalizes IQL into an actor-critic framework and uses diffusion policies for robust policy extraction, outperforming prior offline RL methods.
citing papers explorer
-
EXPO: Stable Reinforcement Learning with Expressive Policies
EXPO stabilizes online RL for expressive policies by training a base policy with imitation and using a lightweight Gaussian edit policy to select higher-value actions on the fly for sampling and TD backups.
-
OGPO: Sample Efficient Full-Finetuning of Generative Control Policies
OGPO is a sample-efficient off-policy method for full finetuning of generative control policies that reaches SOTA on robotic manipulation tasks and can recover from poor behavior-cloning initializations without expert data.
-
IDQL: Implicit Q-Learning as an Actor-Critic Method with Diffusion Policies
IDQL generalizes IQL into an actor-critic framework and uses diffusion policies for robust policy extraction, outperforming prior offline RL methods.