I am a PhD candidate in Physics at the University of California, Berkeley. My research focuses on leveraging advanced machine learning techniques, particularly Generative Models, to analyze particle collider data. My research projects, including Moment Unfolding and Infinite Deconvolution, demonstrate the application of these models to complex datasets, advancing our understanding of particle physics and showcasing their broad applicability.
My academic foundation was laid at Yale, where I earned my BS and MS degrees in Mathematics and Physics with Distinction, completing my studies in an accelerated three-year timeframe. My professional journey includes my tenure as a PhD Research Intern at Microsoft, contributing to the "Theory Explorer" project under Jaron Lanier, and a role at Bridgewater, where I applied my analytical skills to financial datasets.
At Lawrence Berkeley National Lab, advised by Dr. Benjamin Nachman, my work has not only deepened in the realm of high-energy physics through projects like SymmetryGAN but also expanded the potential applications of deep learning in data analysis and insight generation.
Publications
1 Deconvolving Detector Effects for Distribution Moments,
• Krish Desai, Benjamin Nachman, and Jesse Thaler
NeurIPS (ML4PS 2022) 43 – November 28, 2022
2 SymmetryGAN: Symmetry discovery with deep learning
• Krish Desai, Benjamin Nachman, and Jesse Thaler
Phys. Rev. D 105, 096031 – Published 26 May 2022
3 Symmetry Discovery with Deep Learning
• Krish Desai, Benjamin Nachman, and Jesse Thaler
NeurIPS (ML4PS 2021) 117
4 Oblivious points on translation surfaces
• Ian Adelstein, Krish Desai, Anthony Ji, and Grace Zdeblick
Journal of Geometry 113 (6) Published: 16 December 2021
Machine Learning for High Energy Physics