Bridging the sim-to-real gap from the information bottleneck perspective

H He, P Wu, C Bai, H Lai, L Wang, L Pan, X Hu… - arxiv preprint arxiv …, 2023 - arxiv.org
Reinforcement Learning (RL) has recently achieved remarkable success in robotic control.
However, most works in RL operate in simulated environments where privileged knowledge …

Student-Informed Teacher Training

N Messikommer, J **ng, E Aljalbout… - arxiv preprint arxiv …, 2024 - arxiv.org
Imitation learning with a privileged teacher has proven effective for learning complex control
behaviors from high-dimensional inputs, such as images. In this framework, a teacher is …

NPE-DRL: Enhancing Perception Constrained Obstacle Avoidance with Non-Expert Policy Guided Reinforcement Learning

Y Zhang, C Yan, J **ao… - IEEE Transactions on …, 2024 - ieeexplore.ieee.org
Obstacle avoidance under constrained visual perception presents a significant challenge,
requiring rapid detection and decision-making within partially observable environments …

Better than Your Teacher: LLM Agents that learn from Privileged AI Feedback

S Choudhury, P Sodhi - arxiv preprint arxiv:2410.05434, 2024 - arxiv.org
While large language models (LLMs) show impressive decision-making abilities, current
methods lack a mechanism for automatic self-improvement from errors during task …

Learn A Flexible Exploration Model for Parameterized Action Markov Decision Processes

Z Wang, B Wang, M Shao, H Dou, B Tao - arxiv preprint arxiv:2501.02774, 2025 - arxiv.org
Hybrid action models are widely considered an effective approach to reinforcement learning
(RL) modeling. The current mainstream method is to train agents under Parameterized …