Follow me on Twitter, opens an external URL in a new window, where I try to tweet about my current research and about interesting developments in deep learning or related fields. Sometimes I also look at LinkedIn, opens an external URL in a new window.

I was at Max Welling's group at the AMLab Amsterdam, opens an external URL in a new window from January 2021 to January 2022.

I am a Researcher at the newly founded Microsoft Lab in Amsterdam, opens an external URL in a new window from January 2022 onwards.

• Geometric Deep Learning (Graph Networks, Equivariant Networks)   -- on it right now!
• Physics-inspired and physics-informed machine learning -- on it right now!
• Modeling of Dynamical Systems and Partial Differential Equations -- on it right now!
• Hopfield Networks, Transformers and Associative Memories -- some time ago
• Few Shot Learning, Representation Learning -- some time ago
• Reinforcement Learning -- some time ago
• Generative Modeling -- some time ago
• Particle Physics (Higgs boson physics) -- quite some time ago (I can't say I found the Higgs boson, but was one of the guys measuring some of its most important properties).

For a complete list of my publications, have a look at my Google Scholar, opens an external URL in a new window. 

2022

[1] Johannes Brandstetter, Max Welling, Daniel Worrall (2022). Lie Point Symmetry Data Augmentation for Neural PDE Solvers. International Conference on Machine Learning (ICML) 2022  [PDF], opens an external URL in a new window

[2] Johannes Brandstetter*, Daniel Worrall*, Max Welling (2022). Message Passing Neural PDE Solvers. International Conference on Learning Representations 2022. ICLR 2022 SPOTLIGHT PAPER. [PDF], opens an external URL in a new window

2021

[1] Thomas Adler, Manuel Erhard, Mario Krenn, Johannes Brandstetter, Johannes Kofler, and Sepp Hochreiter (2021). Quantum optical experiments modeled by long short-term memory. Photonics 8 (12), 535. [PDF], opens an external URL in a new window

[2] Johannes Brandstetter*, Rob Hesselink*, Elise van der Pol, Erik Bekkers, and Max Welling (2021). Geometric and Physical Quantities Improve E(3) Equivariant Message Passing. International Conference on Learning Representations 2022. ICLR 2022 SPOTLIGHT PAPER. [PDF], opens an external URL in a new window

[3] Andreas Mayr, Sebastian Lehner, Arno Mayrhofer, Christoph Kloss, Sepp Hochreiter, and Johannes Brandstetter (2021). Boundary graph neural networks for 3d simulations. [PDF], opens an external URL in a new window

[4] Andreas Mayr, Sebastian Lehner, Arno Mayrhofer, Christoph Kloss, Sepp Hochreiter, and Johannes Brandstetter (2021). Learning 3D Granular Flow Simulations. International Conference on Learning Representations 2021 SimDL Workshop. [PDF], opens an external URL in a new window

[5] Andreas Mayr, Johannes Brandstetter (2021). Boundary graph neural networks for 3d simulations. Blog post. [Link], opens an external URL in a new window

[6] Daniel Klotz, Frederik Kratzert, Martin Gauch, Alden Keefe Sampson, Johannes Brandstetter, Günter Klambauer, Sepp Hochreiter, and Grey Nearing (2021). Uncertainty estimation with deep learning for rainfall-runoff modelling. Hydrology and Earth System Sciences Discussions 1-32. [PDF], opens an external URL in a new window

2020 

[1] Michael Widrich, Bernhard Schäfl, Hubert Ramsauer, Milena Pavlović, Lukas Gruber, Markus Holzleitner, Johannes Brandstetter, Geir Kjetil Sandve, Victor Greiff, Sepp Hochreiter and Günter Klambauer (2020). Modern Hopfield networks and attention for immune repertoire classification. Conference on Advances in Neural Information Processing Systems 2020. Spotlight Paper. [PDF], opens an external URL in a new window

[2] Hubert Ramsauer, Bernhard Schäfl, Johannes Lehner, Philipp Seidl, Michael Widrich, Lukas Gruber, Markus Holzleitner, Milena Pavlović, Geir Kjetil Sandve, Victor Greiff, David Kreil, Michael Kopp, Günter Klambauer, Johannes Brandstetter, and Sepp Hochreiter (2020). Hopfield networks is all you need. International Conference on Learning Representations 2021. [PDF], opens an external URL in a new window

[3] Vihang P Patil, Markus Hofmarcher, Marius-Constantin Dinu, Matthias Dorfer, Patrick M Blies, Johannes Brandstetter, Jose A Arjona-Medina, and Sepp Hochreiter (2020). Align-RUDDER: Learning From Few Demonstrations by Reward Redistribution. [PDF], opens an external URL in a new window

[4] Thomas Adler*, Johannes Brandstetter*, Michael Widrich, Andreas Mayr, David Kreil, Michael Kopp, Günter Klambauer, and Sepp Hochreiter (2020). Cross-Domain Few-Shot Learing by Representation Fusion. [PDF], opens an external URL in a new window

[5] Markus Holzleitner, Lukas Gruber, Jose Arjona-Medina, Johannes Brandstetter, Sepp Hochreiter (2020). Convergence Proof of Actor-Critic Methods Applied to PPO and RUDDER. Transactions on Large-Scale Data-and Knowledge-Centered Systems XLVIII, 105-130. [PDF], opens an external URL in a new window

[6] Johannes Brandstetter (2020). Hopfield Networks is All You Need. Blog post. [Link, opens an external URL in a new window]

[7] Johannes Brandstetter, Hubert Ramsauer, Markus Holzleitner, Sepp Hochreiter, and Bernhard Schäfl (2020). Looking at the Performer from a Hopfield Point of View. Blog post. [Link, opens an external URL in a new window]

[8] Thomas Adler*, Johannes Brandstetter* (2020). Cross-Domain Hebbian Ensemble Few Shot Learning. Blog post. [Link, opens an external URL in a new window]

2019

[1] Johannes Brandstetter (2019). RUDDER: Return Decomposition for Delayed Rewards. Blog post. [Link, opens an external URL in a new window] 

[2] Jose Arjona-Medina, Michael Gillhofer, Michael Widrich, Thomas Unterthiner, Johannes Brandstetter, and Sepp Hochreiter (2019). RUDDER: Return Decomposition for Delayed Rewards. Advances in Neural Information Processing Systems (13566-13577). [PDF], opens an external URL in a new window

[3] Thomas Adler, Manuel Erhard, Mario Krenn, Johannes Brandstetter, Johannes Kofler, and Sepp Hochreiter (2019). Quantum Optical Experiments Modeled by Long Short-Term Memory. NeurIPS 2019 Workshop: Machine Learning and the Physical Sciences. [PDF], opens an external URL in a new window

[4] Frederik Kratzert, Daniel Klotz, Johannes Brandstetter, Pieter-Jan Hoedt, Grey Nearing, and Sepp Hochreiter (2019). Using LSTMs for climate change assessment studies on droughts and floods. NeurIPS 2019 Workshop: Tackling Climate Change with ML. [PDF], opens an external URL in a new window

[5] Hubert Ramsauer, Johannes Brandstetter, Michael Gillhofer, Bernhard Schäfl, Sepp Hochreiter (2019). A GAN based solver idea for derivative-free optimization problems. NeurIPS 2019 Workshop: Science meets Engineering of Deep Learning.

[6] Michael Gillhofer, Hubert Ramsauer, Johannes Brandstetter, Bernhard Schäfl, Sepp Hochreiter (2019). A GAN based solver of black-box inverse problems. NeurIPS 2019 Workshop: Solving inverse problems with deep networks.

•  Analysis of Integral Probability Metrics for Generative Adversarial Networks, Lukas Gruber (2019)
•  Semantic Segmentation of Different Polymers in Hyperspectral Images using Deep Neural Networks, Kristina Duswald (2022)

• 2018: Exercises in Machine Learning: Supervised Techniques, Lecturer, JKU Linz
• 2018: Lecture in Numerical and Symbolic Methods for Bioinformatics, Lecturer, JKU Linz
• 2019: Exercises in Machine Learning: Unsupervised Techniques, Lecturer, JKU Linz
• 2019: Exercises in Theoretical Concepts of Machine Learning, Lecturer, JKU Linz
• 2019: Lecture in Machine Learning: Supervised Techniques, Lecturer, JKU Linz
• 2019: Exercises in Machine Learning: Supervised Techniques, Lecturer, JKU Linz
• 2019: Lecture in Hands-on AI I, Lecturer, JKU Linz
• 2019: Lecture in Programming in Python I, Co-Lecturer, JKU Linz
• 2020: Lecture in Hands-on AI II, Lecturer, JKU Linz
• 2020: Lecture in Deep Learning and Neural Nets II, Co-Lecturer, JKU Linz
• 2020: Lecture in Hands-on AI I, Lecturer, JKU Linz
• 2020: Exercises in Hands-on AI I, Lecturer, JKU Linz

• ÖAW AI Summer School 2019, Deep Learning Lecture series. (Homepage), opens an external URL in a new window

• Blog post to the paper RUDDER: Return Decomposition for Delayed Rewards, opens an external URL in a new window.
• Blog post to the paper Hopfield Networks is All You Need, opens an external URL in a new window.
• Blog post to the paper Align-RUDDER: Learning from Few Demonstrations by Reward Redistribution, opens an external URL in a new window.
• Blog post to the paper Cross-Domain Few-Shot Learning by Representation Fusion, opens an external URL in a new window.
• Blog post Looking at the Performer from a Hopfield point of view, opens an external URL in a new window.
• Blog post to the paper Boundary Graph Neural Networks for 3D Simulations, opens an external URL in a new window.

• Born September 25, 1989
• BSc in Technical Physics (TU Wien), July 2012
• MSc in Technical Physics (TU Wien), November 2014
• PhD in High Energy Physics (CERN, Hephy Vienna), May 2018 (PhD Thesis, opens an external URL in a new window)
• Post-Doc at the Institute for Machine Learning (JKU Linz), July 2018 - July 2020
• Assistant Professor (Tenure Track, JKU Linz), August 2020 - ongoing
• ELLIS Member, opens an external URL in a new window
• ELLIS Research at the University of Amsterdam (Max Welling group), January 2021 - January 2022
• Researcher at the newly founded Microsoft Lab in Amsterdam, opens an external URL in a new window, January 2022 onwards