Offline reinforcement learning requires reconciling two conflicting aims: learning a policy that
improves over the behavior policy that collected the dataset, while at the same time …

Offline reinforcement learning (RL) defines the task of learning from a fixed batch of data.
Due to errors in value estimation from out-of-distribution actions, most offline RL algorithms …

Reinforcement learning (RL) is typically viewed as the problem of estimating single-step
policies (for model-free RL) or single-step models (for model-based RL), leveraging the …

Abstract Model-based reinforcement learning (RL) algorithms, which learn a dynamics
model from logged experience and perform conservative planning under the learned model …

Imitating human demonstrations is a promising approach to endow robots with various
manipulation capabilities. While recent advances have been made in imitation learning and …

Offline (or batch) reinforcement learning (RL) algorithms seek to learn an optimal policy from
a fixed dataset without active data collection. Based on the composition of the offline dataset …

A compelling use case of offline reinforcement learning (RL) is to obtain a policy initialization
from existing datasets followed by fast online fine-tuning with limited interaction. However …

Effective offline RL methods require properly handling out-of-distribution actions. Implicit Q-
learning (IQL) addresses this by training a Q-function using only dataset actions through a …

Offline reinforcement learning (RL) defines the task of learning from a static logged dataset
without continually interacting with the environment. The distribution shift between the …

Recent advance in deep offline reinforcement learning (RL) has made it possible to train
strong robotic agents from offline datasets. However, depending on the quality of the trained …