Data-driven and physics-informed probabilistic machine learning
Bayesian Methods
Uncertainty Quantification and Propagation
Inversion and stochastic optimization in the context of PDEs
Teaching
Uncertainty Modeling in Engineering (SS 18 - WS22)
Probability Theory and Uncertainty Quantification (WS 17/18 - WS 22)
Uncertainty Quantification in Mechanical Engineering (SS 2017)
Bayesian Strategies for Inverse Problems (SS 2017)
Publications
M. Rixner and P.S. Koutsourelakis, Self-supervised optimization of random material microstructures in the small-data regime, Nature Partner Journal of Computational Materials (2022)
M. Rixner and P.S. Koutsourelakis, A probabilistic generative model for semi-supervised training of coarse-grained surrogates and enforcing physical constraints through virtual observables, Journal of Computational Physics (2021)
Conference Contributions
M. Rixner, P.S. Koutsourelakis, Data-efficient, adaptive learning in optimization under uncertainty: applications in materials' design, SIAM Conference on Uncertainty Quantification, Atlanta, USA, 2022
M. Rixner, P.S. Koutsourelakis, Incorporating Physics-Based, Inductive Bias in Deep, Probabilistic Surrogates of PDEs with High-Dimensional Inputs, SIAM Conference on Computational Science and Engineering, Virtual, 2021
M. Rixner, P.S. Koutsourelakis, Accelerating Physics-constrained Bayesian Inverse Problems using Inaccurate Models and Data-driven Learning, SIAM Conference on Computational Science and Engineering, Spokane, USA, 2019
M. Rixner, P.S. Koutsourelakis, Solution of PDE Constrained Inverse Problems from a Machine Learning Perspective, World Congress on Computational Mechanics, New York City, USA, 2018
M. Rixner, P.S. Koutsourelakis, Incorporating Epistemic Uncertainty from Lower-fidelity Models in Bayesian Inverse Problems, SIAM Conference on Uncertainty Quantification, Los Angeles, USA, 2018
M. Rixner, P.S. Koutsourelakis, Beyond Black-boxes in Model-based Bayesian inverse Problems, SIAM Conference on Uncertainty Quantification, Los Angeles, USA, 2018
M. Rixner, P.S. Koutsourelakis, Tutorial on Bayesian Multi-Level Monte Carlo, 46th SpeedUp Workshop on Uncertainty Quantification and HPC, Bern, 2017
M. Rixner, P.S. Koutsourelakis, Bayesian, Multi-Fidelity Optimization under Uncertainty, SIAM Conference on Computation in Science and Engineering, Atlanta, USA, 2017
Supervised Student Projects
Medical Image Synthesis Using Generative Adversarial Networks (Semester's Thesis), 2020 (in collaboration with Hongwei Li at the Chair for Computer Aided Medical Procedures & Augmented Reality)
Active Machine Learning using Gaussian Processes (Bachelor's Thesis), 2018
Development of an algorithm for determining the severity of injury using accident data (Master's Thesis), 2018 (in collaboration with BMW)
A Bayesian Multi-Fidelity Approach for Inverse Problems (Master's Thesis), 2018
