Felix Berkenkamp

PhD Student
External Website
befelix@inf.ethz.ch
CAB G 66.2
+41 44 632 33 08
LinkedIn
My main research interest lies at the intersection of machine learning, control theory, and robotics. In particular, I focus on the questions of how robots can safely explore unknown environments and how they can improve their controllers without failures. This currently requires new algorithms that combine ideas from Bayesian optimization, (robust) control theory and reinforcement learning. More information here.

Publications

Deprecated: Function get_magic_quotes_runtime() is deprecated in /var/www/html/site-las/biblib.php on line 310

2021
Bayesian Optimization with Safety Constraints: Safe and Automatic Parameter Tuning in Robotics F. Berkenkamp, A. Krause, A. P. Schoellig In Machine Learning Journal, Special Issue on Robust Machine Learning, 2021 [bibtex] [abstract] [pdf] [code] [video] [doi]
2020
Efficient Model-Based Reinforcement Learning through Optimistic Policy Search and Planning S. Curi, F. Berkenkamp, A. Krause In Proc. Neural Information Processing Systems (NeurIPS), 2020 Spotlight presentation [bibtex] [abstract] [pdf] [poster] [code]
Structured Variational Inference in Partially Observable Unstable Gaussian Process State Space Models S. Curi, S. Melchior, F. Berkenkamp, A. Krause In Proc. Conference on Learning for Dynamics and Control (L4DC), 2020 [bibtex] [abstract] [pdf]
2019
Safe Exploration for Interactive Machine Learning M. Turchetta, F. Berkenkamp, A. Krause In Proc. Neural Information Processing Systems (NeurIPS), 2019 [bibtex] [abstract] [pdf]
Information-Directed Exploration for Deep Reinforcement Learning N. Nikolov, J. Kirschner, F. Berkenkamp, A. Krause In Proc. International Conference on Learning Representations (ICLR), 2019 [bibtex] [abstract] [pdf] [poster]
No-Regret Bayesian Optimization with Unknown Hyperparameters F. Berkenkamp, A. P. Schoellig, A. Krause In Journal of Machine Learning Research (JMLR), volume 20, 2019 [bibtex] [abstract] [pdf]
2018
Learning-based Model Predictive Control for Safe Exploration T. Koller, F. Berkenkamp, M. Turchetta, A. Krause In Proc. of the IEEE Conference on Decision and Control (CDC), 2018 [bibtex] [abstract] [pdf] [code]
The Lyapunov Neural Network: Adaptive Stability Certification for Safe Learning of Dynamical Systems S. M. Richards, F. Berkenkamp, A. Krause In Proceedings of The 2nd Conference on Robot Learning, PMLR, volume 87, 2018 Oral presentation [bibtex] [abstract] [pdf] [code] [video]
2017
Safe Model-based Reinforcement Learning with Stability Guarantees F. Berkenkamp, M. Turchetta, A. P. Schoellig, A. Krause In Proc. Neural Information Processing Systems (NeurIPS), 2017 [bibtex] [abstract] [pdf] [code] [video]
Virtual vs. Real: Trading Off Simulations and Physical Experiments in Reinforcement Learning with Bayesian Optimization A. Marco, F. Berkenkamp, P. Hennig, A. P. Schoellig, A. Krause, S. Schaal, S. Trimpe In Proc. of the International Conference on Robotics and Automation (ICRA), 2017 [bibtex] [abstract] [pdf]
2016
Safe Exploration in Finite Markov Decision Processes with Gaussian Processes M. Turchetta, F. Berkenkamp, A. Krause In Proc. Neural Information Processing Systems (NeurIPS), 2016 [bibtex] [abstract] [pdf] [code] [video]
Safe Learning of Regions of Attraction for Uncertain, Nonlinear Systems with Gaussian Processes F. Berkenkamp, R. Moriconi, A. P. Schoellig, A. Krause In Proc. of the IEEE Conference on Decision and Control (CDC), 2016 [bibtex] [abstract] [pdf] [talk] [code]
Bayesian Optimization for Maximum Power Point Tracking in Photovoltaic Power Plants H. Abdelrahman, F. Berkenkamp, J. Poland, A. Krause In Proc. European Control Conference (ECC), 2016 Best Application Paper Award [bibtex] [abstract] [pdf]
Safe Controller Optimization for Quadrotors with Gaussian Processes F. Berkenkamp, A. P. Schoellig, A. Krause In Proc. of the International Conference on Robotics and Automation (ICRA), 2016 [bibtex] [abstract] [pdf] [code] [video]

Felix Berkenkamp

Publications

2021

2020

2019

2018

2017

2016