about me

I am a research fellow at IAIFI.

Previously I was at Perimeter Institute for Theoretical Physics and the University of Waterloo where I did my PhD with Roger Melko at the PIQuIL.

My research is on the intersection between theoretical quantum many-body physics and machine learning: I apply machine learning as a computational tool to solve problems in quantum physics, and I use approaches from theoretical physics to understand deep learning. My goal is to develop a theoretical foundation of deep learning which would in turn allow to expand its applicability in physics. I am particularly interested in approaches involving information theory, statistical learning theory, and statistical physics. Several of my recent projects on the machine-learning side focus on adversarial attacks and model robustness.

Previously, I completed the Perimeter Scholars International master's program (2017), a MSc in Theoretical Physics with focus on computational approaches to quantum many-body systems (2016), and a BSc in Biophysics (2014) at the Goethe University in Frankfurt, Germany.


October 2021 our work on RBM pruning is out: "Pruning a restricted Boltzmann machine for quantum state reconstruction"
September 2021 I am super excited to be joining IAIFI as a fellow!
July 2021 defending my thesis @ UW!
July 2021 I'm giving an invited talk at the workshop Sparsity in Neural Networks
May 2021 I gave a talk about our work on wide and sparse networks @ Physics ∩ ML
May 2021 presenting our work "Are wider nets better given the same number of parameters?" @ ICLR 2021
November 2019 - May 2020 I'm out for a research internship with Blueshift @ Alphabet Inc., working on wide and sparse neural networks
August 2019 attending the Waterloo ML + Security + Verification Workshop @ UW and presenting our work on Batch Norm
August 2019 our scaling study on RBMs is out: The learnability scaling of quantum states: restricted Boltzmann machines
July 2019 attending the conference Machine Learning for Quantum Design @ Perimeter
June 2019 QuCumber: wavefunction reconstruction with neural networks, a software product from PIQUIL, now published on SciPost
June 2019 presenting our recent work on Batch Norm @ ICML 2019 workshop "Deep Phenomena"
May 2019 new work with Angus Galloway: Batch Normalization is a Cause of Adversarial Vulnerability
May 2019 I'm honoured to receive the NSERC Gilles Brassard Doctoral Prize for Interdisciplinary Research!
Mar 2019 thrilled to join the 2019 Vector Institute Postgraduate Affiliate Program!
Feb 2019 I'm giving a talk @ Caltech Theoretical Condensed Matter Physics department
Jan 2019 attending the program "Machine Learning for Quantum Many-Body Physics" @ KITP Santa Barbara