Background

Debajyoti Sengupta (Deb) is a postdoctoral researcher at EPFL’s MIPLab, working on generative modeling and multimodal machine learning for biomedical applications. He defended his Ph.D. in Physics and Computer Science in October 2024 in the group of Prof. Tobias Golling at the Particle Physics Department (DPNC), where he was a member of the RODEM project (Robust Deep Density Models for High‑Energy Physics and Solar Physics) and a qualified author of the ATLAS Collaboration at CERN.

Debajyoti’s doctoral and early postdoctoral research focused on weakly supervised anomaly detection, diffusion‑based generative modelling, and representation learning via contrastive methods, with applications in high‑energy physics and astrophysics. His prior work included GAN‑based calorimeter simulations and data analysis of Drell–Yan processes for a rediscovery of the Z boson.

At EPFL, his current research extends these methodological foundations to neuro-technology and medical AI, with a particular interest in inverse problems such as M/EEG source localisation, fMRI deconvolution, and dynamics discovery in neuroimaging data. By combining rigorous machine learning approaches with translational impact, his work aims to bridge fundamental advances in AI with practical deployment in healthcare and neuroscience.