SOE: Sample-Efficient Robot Policy Self-Improvement via On-Manifold Exploration
This addresses the challenge of safe and efficient policy self-improvement for robotics, offering a plug-in solution that enhances exploration without degrading base performance, though it is incremental as it builds on existing policy models.
The paper tackles the problem of insufficient exploration in robot policies due to action mode collapse by proposing SOE, a framework that constrains exploration to a manifold of valid actions, resulting in higher task success rates, smoother exploration, and superior sample efficiency in robotic manipulation tasks.
Intelligent agents progress by continually refining their capabilities through actively exploring environments. Yet robot policies often lack sufficient exploration capability due to action mode collapse. Existing methods that encourage exploration typically rely on random perturbations, which are unsafe and induce unstable, erratic behaviors, thereby limiting their effectiveness. We propose Self-Improvement via On-Manifold Exploration (SOE), a framework that enhances policy exploration and improvement in robotic manipulation. SOE learns a compact latent representation of task-relevant factors and constrains exploration to the manifold of valid actions, ensuring safety, diversity, and effectiveness. It can be seamlessly integrated with arbitrary policy models as a plug-in module, augmenting exploration without degrading the base policy performance. Moreover, the structured latent space enables human-guided exploration, further improving efficiency and controllability. Extensive experiments in both simulation and real-world tasks demonstrate that SOE consistently outperforms prior methods, achieving higher task success rates, smoother and safer exploration, and superior sample efficiency. These results establish on-manifold exploration as a principled approach to sample-efficient policy self-improvement. Project website: https://ericjin2002.github.io/SOE