Share This Author
Safe Model-based Reinforcement Learning with Stability Guarantees
This paper presents a learning algorithm that explicitly considers safety, defined in terms of stability guarantees, and extends control-theoretic results on Lyapunov stability verification and shows how to use statistical models of the dynamics to obtain high-performance control policies with provable stability certificates.
Learning-Based Model Predictive Control for Safe Exploration
- T. Koller, Felix Berkenkamp, M. Turchetta, A. Krause
- Computer ScienceIEEE Conference on Decision and Control
- 22 March 2018
This paper presents a learning-based model predictive control scheme that can provide provable high-probability safety guarantees and exploits regularity assumptions on the dynamics in terms of a Gaussian process prior to construct provably accurate confidence intervals on predicted trajectories.
Safe controller optimization for quadrotors with Gaussian processes
- Felix Berkenkamp, Angela P. Schoellig, Andreas Krause
- Engineering, Computer ScienceIEEE International Conference on Robotics and…
- 3 September 2015
Experimental results on a quadrotor vehicle indicate that the proposed SafeOpt algorithm enables fast, automatic, and safe optimization of controller parameters without human intervention.
The Lyapunov Neural Network: Adaptive Stability Certification for Safe Learning of Dynamic Systems
- Spencer M. Richards, Felix Berkenkamp, Andreas Krause
- Computer ScienceConference on Robot Learning
- 2 August 2018
A method to learn accurate safety certificates for nonlinear, closed-loop dynamical systems by constructing a neural network Lyapunov function and a training algorithm that adapts it to the shape of the largest safe region in the state space.
Keep Doing What Worked: Behavioral Modelling Priors for Offline Reinforcement Learning
- Noah Siegel, J. T. Springenberg, Martin A. Riedmiller
- Computer ScienceInternational Conference on Learning…
- 19 February 2020
This paper admits the use of data generated by arbitrary behavior policies and uses a learned prior -- the advantage-weighted behavior model (ABM) -- to bias the RL policy towards actions that have previously been executed and are likely to be successful on the new task.
Bayesian Optimization with Safety Constraints: Safe and Automatic Parameter Tuning in Robotics
- Felix Berkenkamp, Andreas Krause, Angela P. Schoellig
- Computer ScienceMachine-mediated learning
- 14 February 2016
A generalized algorithm that allows for multiple safety constraints separate from the objective is presented, which enables fast, automatic, and safe optimization of tuning parameters in experiments on a quadrotor vehicle.
Safe Exploration in Finite Markov Decision Processes with Gaussian Processes
A novel algorithm is developed and proved that it is able to completely explore the safely reachable part of the MDP without violating the safety constraint, and is demonstrated on digital terrain models for the task of exploring an unknown map with a rover.
Safe learning of regions of attraction for uncertain, nonlinear systems with Gaussian processes
- Felix Berkenkamp, R. Moriconi, Angela P. Schoellig, Andreas Krause
- Mathematics, Computer ScienceIEEE Conference on Decision and Control
- 15 March 2016
This paper considers an approach that learns the ROA from experiments on a real system, without ever leaving the true ROA and, thus, without risking safety-critical failures.
Verifying Controllers Against Adversarial Examples with Bayesian Optimization
- S. Ghosh, Felix Berkenkamp, G. Ranade, S. Qadeer, Ashish Kapoor
- Computer ScienceIEEE International Conference on Robotics and…
- 23 February 2018
This paper presents an active-testing framework based on Bayesian Optimization that specifies safety constraints using logic and exploit structure in the problem in order to test the system for adversarial counter examples that violate the safety specifications.
Efficient Model-Based Reinforcement Learning through Optimistic Policy Search and Planning
- Sebastian Curi, Felix Berkenkamp, A. Krause
- Computer ScienceNeural Information Processing Systems
- 15 June 2020
This paper proposes a practical optimistic-exploration algorithm, which enlarges the input space with hallucinated inputs that can exert as much control as the epistemic uncertainty in the model affords, and shows how optimistic exploration can be easily combined with state-of-the-art reinforcement learning algorithms and different probabilistic models.