About me
I am a postdoctoral researcher at ETH Zurich working in the Learning & Adaptive Systems group of the Institute for Machine Learning advised by Prof. Andreas Krause. My research focuses on ensuring the safety of unknown systems through a combination of techniques from model-based control and supervised machine learning. My goal is to develop theoretically rigorous and computationally efficient methods, whose effectiveness is demonstrated in experiments with robots such as manipulators and underwater vehicles.
Before Joining ETH Zurich in October 2023, I obtained a PhD, Master’s, and Bachelor’s degree in Electrical and Computer Engineering at the Technical University of Munich, Germany, in 2023, 2018, and 2015, respectively. During my time as a PhD student, I worked under the supervision of Prof. Sandra Hirche on learning-based control partially funded by a scholarship of the German Academic Scholarship Foundation. In 2016, I spent a semester abroad at University of Illinois at Urbana-Champaign.
Student Supervision I am open to supervising ambitious and talented Master’s and Bachelor’s students for their thesis. If you want to work with me, please send me an email describing your area of interest. Please also attach your CV and up-to-date transcripts.
News
| 03/24 | I have given a presentation on “Safe Online Learning Control using Gaussian Processes” in the Systems and Control Division at Uppsala University. |
| 02/24 | Our paper “Learning-Based Optimal Control with Performance Guarantees for Unknown Systems with Latent States” has been accepted to the European Control Conference. Check out the paper here! |
| 02/24 | I have given a remote seminar at the National University of Singapore. |
| 01/24 | I have given a virtual guest lecture at UC San Diego on Safe Model-Based Control with Gaussian Processes. You can watch the talk online here. |
| 01/24 | Our paper “Cooperative Learning with Gaussian Processes for Euler-Lagrange Systems Tracking Control under Switching Topologies” has been accepted to the American Control Conference! |
| 12/23 | I attended IEEE Conference on Decision and Control in Singapore and presented our paper “Risk-Sensitive Inhibitory Control for Safe Reinforcement Learning”. Check out the paper here! |
| 12/23 | Petar and Max presented our work on Koopman kernel regression at the Conference on Neural Information Processing Systems (NeurIPS) in New Orleans. You can find the paper here. |
| 11/23 | I received a top reviewer award at NeurIPS 2023. |
| 11/23 | I gave a presentation at the University of Groningen entitled “Safe Intelligent Systems for the Physical World” |
| 10/23 | I gave an online seminar the University of Groningen with the topic “Safe and Computationally Efficient Online Learning Control using Gaussian Processes” |
| 10/23 | I have started my new positition as postdoctoral researcher at the Learning & Adaptive Systems Group of ETH Zurich supervised by Prof. Andreas Krause |