Safe Reinforcement Learning via Confidence-Based Filters

by   Sebastian Curi, et al.

Ensuring safety is a crucial challenge when deploying reinforcement learning (RL) to real-world systems. We develop confidence-based safety filters, a control-theoretic approach for certifying state safety constraints for nominal policies learned via standard RL techniques, based on probabilistic dynamics models. Our approach is based on a reformulation of state constraints in terms of cost functions, reducing safety verification to a standard RL task. By exploiting the concept of hallucinating inputs, we extend this formulation to determine a "backup" policy that is safe for the unknown system with high probability. Finally, the nominal policy is minimally adjusted at every time step during a roll-out towards the backup policy, such that safe recovery can be guaranteed afterwards. We provide formal safety guarantees, and empirically demonstrate the effectiveness of our approach.


page 1

page 2

page 3

page 4


Safe Reinforcement Learning via Probabilistic Logic Shields

Safe Reinforcement learning (Safe RL) aims at learning optimal policies ...

Safe Distributional Reinforcement Learning

Safety in reinforcement learning (RL) is a key property in both training...

Safe reinforcement learning for probabilistic reachability and safety specifications: A Lyapunov-based approach

Emerging applications in robotics and autonomous systems, such as autono...

Log Barriers for Safe Black-box Optimization with Application to Safe Reinforcement Learning

Optimizing noisy functions online, when evaluating the objective require...

Safe Exploration in Continuous Action Spaces

We address the problem of deploying a reinforcement learning (RL) agent ...

Conservative Safety Critics for Exploration

Safe exploration presents a major challenge in reinforcement learning (R...

Value Functions are Control Barrier Functions: Verification of Safe Policies using Control Theory

Guaranteeing safe behaviour of reinforcement learning (RL) policies pose...

Please sign up or login with your details

Forgot password? Click here to reset