SOHAM DE

Research Scientist
DeepMind
London, UK

Contact Address:
R7, 14-18 Handyside St,
London, UK, N1C 4DN

Email / Scholar / Twitter

I am a research scientist at DeepMind in London, where I work on better understanding and improving large-scale deep learning. I currently focus on topics in optimization and learning dynamics, efficient training and inference, and scaling principles. I am also interested in robustness and privacy of deep learning models. Feel free to drop me a line if you want to chat about any of these topics.

I completed my PhD in 2018 at the University of Maryland, where I was advised by Dana Nau and Tom Goldstein. During my PhD, I worked on fast stochastic optimization algorithms for machine learning problems. I also worked on game-theoretic models of the evolution of human behavioral norms collaborating with Michele Gelfand. I completed my undergraduate degree from Jadavpur University, Kolkata, India in 2013. I have previously interned at Toyota Technological Institute at Chicago (TTIC), IBM Research Almaden and DeepMind.

RECENT TALKS


SELECTED PAPERS

Full List: Google Scholar


Last Updated: November 28, 2024