ML

Interested in precision measurements of the Higgs boson, as well as searches for new Higgs-like particles to explain mysteries such as baryon asymmetry.

Leading the effort on several state-of-the-art generative models to accelerate LHC simulations. Developing as well validation and benchmarking schemes in order to bring them into CMS. Published at NeurIPS, PRD, and more.

Developed graph-based and Lorentz-equivariant models to better suit our high energy physics data. Our Lorentz-group autoencoder (LGAE) outperforms graph and convolutional networks on jet compression and anomaly detection tasks. Latest work published at EPJC.

Developed a library for convenient access to jet datasets, and other utilities, to increase accessibility and reproducibility in ML in particle physics. >35,000 downloads as of September 2023, used in several ML and particle physics projects.

Developed several approaches for finding rare particle collisions, including GNNs, Lorentz-equivariant networks, and multi-variate goodness-of-fit tests.

Interpreting results of machine learning models for reconstruction and jet classification using explainable AI techniques.

Fun project to gain experience with RNNs and Attention. I achieved a 71% testing accuracy in predicting the outcome of European football matches.

A novel flow matching technique for inferring physical trajectories from observations snapshots, based on a physics-inspired modification of dynamic optimal transport. SOTA results on biological, meteorogical, and oceanographic datasets.

A novel physics-informed self-supervised pretraining strategy for foundation models in physics. SOTA results on transfer learning between simulations and real data.