State-Conditional Adversarial Learning: An Off-Policy Visual Domain Transfer Method for End-to-End Imitation Learning
By: Yuxiang Liu, Shengfan Cao
We study visual domain transfer for end-to-end imitation learning in a realistic and challenging setting where target-domain data are strictly off-policy, expert-free, and scarce. We first provide a theoretical analysis showing that the target-domain imitation loss can be upper bounded by the source-domain loss plus a state-conditional latent KL divergence between source and target observation models. Guided by this result, we propose State- Conditional Adversarial Learning, an off-policy adversarial framework that aligns latent distributions conditioned on system state using a discriminator-based estimator of the conditional KL term. Experiments on visually diverse autonomous driving environments built on the BARC-CARLA simulator demonstrate that SCAL achieves robust transfer and strong sample efficiency.
Similar Papers
Prompt-Driven Domain Adaptation for End-to-End Autonomous Driving via In-Context RL
Robotics
Teaches self-driving cars to drive in bad weather.
Enabling Off-Policy Imitation Learning with Deep Actor Critic Stabilization
Machine Learning (CS)
Teaches robots to copy experts faster.
A Graphical Approach to State Variable Selection in Off-policy Learning
Methodology
Helps computers learn from past choices.