2 papers
cs.LG2026
Policy as Data: Replay-Based Policy Dual Averaging via Advantage Regression
Nianli Peng, Geoffrey J. Gordon, Kianté Brantley
Actor-critic methods reuse past experience to improve sample efficiency. However, historical data are typically regarded as off-policy samples for the current policy-improvement up…
cs.LG2024
LICORICE: Label-Efficient Concept-Based Interpretable Reinforcement Learning
Zhuorui Ye, Stephanie Milani, Geoffrey J. Gordon +1
Recent advances in reinforcement learning (RL) have predominantly leveraged neural network policies for decision-making, yet these models often lack interpretability, posing challe…