Simon Buchholz
I am a postdoctoral researcher at the Max Planck Institute for Intelligent Systems in the Empirical Inference Department, and I am supported by the Tübingen AI Center.
My main research interests are representation learning and causality. I did my PhD in mathematics at the University of Bonn, where I worked at the intersection of probability theory and statistical mechanics.
My email is sbuchholz at tue dot mpg dot de
Teaching
Summer term 2024: Representations in Generative AI, block seminar at ETH together with Bernhard Schölkopf, Michel Besserve, Zhijing Jin
Summer term 2024: Mathematical Foundations of Machine Learning
Publications and Preprints
Machine learning related:
Simon Buchholz*, Goutham Rajendran*, Elan Rosenfeld, Bryon Aragam, Bernhard Schölkopf, and Pradeep Ravikumar, Learning Linear Causal Representations from Interventions under General Nonlinear Mixing, NeurIPS (oral), 2023. (arxiv)
Jonas Wildberger*, Maximilian Dax*, Simon Buchholz*, Stephen Green, Jakob Macke, and Bernhard Schölkopf, Flow Matching for Scalable Simulation-Based Inference, NeurIPS,
also at ICML 2023 Workshop on Structured Probabilistic Inference & Generative Modeling and 2nd ICML Workshop on Machine Learning for Astrophysics, 2023. (arxiv)
Wendong Liang, Armin Kekic, Julius von Kügelgen, Simon Buchholz, Michel Besserve, Luigi Gresele*, and Bernhard Schölkopf*, Causal Component Analysis, NeurIPS 2023. (arxiv)
Junhyung Park, Simon Buchholz, Bernhard Schölkopf, and Krikamol Muandet, A Measure-Theoretic Axiomatisation of Causality, NeurIPS (oral), 2023. (arxiv)
Simon Buchholz, Jonas Kübler, and Bernhard Schölkopf, Multi Armed Bandits and Quantum Channel Oracles, arXiv preprint arXiv:2301.08544, 2023.
Simon Buchholz, Some Remarks on Identifiability of Independent Component Analysis in restricted Function Classes, Transactions on Machine Learning Research, 2023. (link)
Simon Buchholz, Michel Besserve, and Bernhard Schölkopf, Function Classes for Identifiable Nonlinear Independent Component Analysis, NeurIPS, 2022. (link)
Jonas Kübler, Vincent Stimper, Simon Buchholz, Krikamol Muandet, and Bernhard Schölkopf, AutoML Two-Sample Test, NeurIPS, 2022. (link)
Simon Buchholz, Kernel interpolation in Sobolev spaces is not consistent in low dimensions, COLT, 2022. (link)
Jonas Kübler*, Simon Buchholz*, and Bernhard Schölkopf, The Inductive Bias of Quantum Kernels, NeurIPS, 2021. (link)
Math related:
Simon Buchholz and Codina Cotar, Aizenman-Wehr Argument for a Class of disordered Gradient Models, arXiv preprint arXiv:2309.12799, 2023.
Stefan Adams, Simon Buchholz, Roman Kotecky, and Stefan Müller, Cauchy-Born Rule from Microscopic Models with non-convex Potentials, arXiv preprint arXiv:1910.13564, 2019.
Simon Buchholz, Phase Transitions for a Class of Gradient Fields, Probability Theory and Related Fields, 179(3):969--1022, 2021. (link)
Simon Buchholz, Jean-Dominique Deuschel, Noemi Kurt, and Florian Schweiger, Probability to be positive for the Membrane Model in Dimensions 2 and 3, Electronic Communications in Probability, 24:1 -- 14, 2019. (link)
Simon Buchholz, Finite range decomposition for Gaussian measures with improved regularity, Journal of Functional Analysis, 275(7):1674--1711, 2018. (link)
Simon Buchholz, Chiara Saffirio, and Benjamin Schlein, Multivariate Central Limit Theorem in Quantum Dynamics, Journal of Statistical Physics, 154(1):113--152, 2014. (link)