ISAACS: Iterative Soft Adversarial Actor-Critic for Safety

no code implementations • 6 Dec 2022 • Kai-Chieh Hsu, Duy Phuong Nguyen, Jaime Fernández Fisac

The deployment of robots in uncontrolled environments requires them to operate robustly under previously unseen scenarios, like irregular terrain and wind conditions.

Enhancing Heterogeneous Federated Learning with Knowledge Extraction and Multi-Model Fusion

no code implementations • 16 Aug 2022 • Duy Phuong Nguyen, Sixing Yu, J. Pablo Muñoz, Ali Jannesari

This method allows efficient multi-model knowledge fusion and the deployment of resource-aware models while preserving model heterogeneity.

Federated Learning Knowledge Distillation

Sim-to-Lab-to-Real: Safe Reinforcement Learning with Shielding and Generalization Guarantees

no code implementations • 20 Jan 2022 • Kai-Chieh Hsu, Allen Z. Ren, Duy Phuong Nguyen, Anirudha Majumdar, Jaime F. Fisac

To improve safety, we apply a dual policy setup where a performance policy is trained using the cumulative task reward and a backup (safety) policy is trained by solving the Safety Bellman Equation based on Hamilton-Jacobi (HJ) reachability analysis.

reinforcement-learning Reinforcement Learning (RL) +1