Picture for Jonathan Margoliash

Jonathan Margoliash

Efficient and Safe Exploration in Deterministic Markov Decision Processes with Unknown Transition Models

Add code
Apr 01, 2019
Erdem Bıyık, Jonathan Margoliash, Shahrouz Ryan Alimo, Dorsa Sadigh
Figure 1 for Efficient and Safe Exploration in Deterministic Markov Decision Processes with Unknown Transition Models
Figure 2 for Efficient and Safe Exploration in Deterministic Markov Decision Processes with Unknown Transition Models
Figure 3 for Efficient and Safe Exploration in Deterministic Markov Decision Processes with Unknown Transition Models
Figure 4 for Efficient and Safe Exploration in Deterministic Markov Decision Processes with Unknown Transition Models

We propose a safe exploration algorithm for deterministic Markov Decision Processes with unknown transition models. Our algorithm guarantees safety by leveraging Lipschitz-continuity to ensure that no unsafe states are visited during exploration. Unlike many other existing techniques, the provided safety guarantee is deterministic. Our algorithm is optimized to reduce the number of actions needed for exploring the safe space. We demonstrate the performance of our algorithm in comparison with baseline methods in simulation on navigation tasks.

* Proceedings of the American Control Conference (ACC), July 2019. The first two authors have equal contribution 
Viaarxiv icon
Github IconAccess Paper or Ask Questions
Twitter IconFacebook IconLinkedin IconWhatsapp IconMessenger Icon