S. Karthik Mukkavilli

Dr S. Karthik Mukkavilli is a research scientist working with Mila’s ‘Visualizing Climate Change’ team led by the Scientific Director, Prof. Yoshua Bengio, who recently won the Turing Award, considered the ‘Nobel Prize of computing’. The team’s initial work was featured in MIT tech review. In this team, Dr. Mukkavilli is supervising research students and interns focusing on extreme climate prediction and integrating economics/policy considerations with physics, statistics, machine learning and reinforcement learning approaches. He also teaches and has lectured in the AI4Good summer program. He was one of the co-authors on the manuscript, ‘Tackling Climate Change with Machine Learning’ by Rolnick et al. along with the AI ​​experts, ‘Andrew Ng, Demis Hassabis, Jennifer Chayes and Yoshua Bengio’, featured in MIT tech review .

Research Students/Visiting Interns working on Integrated Climate Extremes within ‘Visualizing Climate Change’ team

Shivam Patel: Develops multi-agent integrated assessment model with reinforcement learning to evaluate economic impacts on climate. From Gujarat, India and previously at Caltech.

Yimeng Min: Predicts changes in the cryosphere, glacier retreat from black carbon and contribution to floods with satellite, physics, deep learning and generative models. From Nanjing, China and previously at Sargent Group in Toronto.

Vitória Barin Pacela: Predicts stream flow contribution to floods with recurrent NNs. From Brazil + Finland and previously at CERN in high energy physics and deep learning / GANs .

Ata Madanchi: Develops a stochastic prediction system for precipitation and hydrodynamic models of floods. From Iran and previously, physics at McGill University.

If you are interested in physics/AI topics related to Integrated Climate Extremes work within visualizing team’s project, as a visiting student, intern, graduate student, or collaborator, please reach out to me. Thank you!

Background:

In addition to Mila, Dr. Mukkavilli has been affiliated as a postdoc in the School of Computer Science at McGill University with Gregory Dudek and David Meger and leads a project called EnviroNet: a data repository of ImageNet analogs for planetary challenges to benchmark machine learning progress in geosciences. EnviroNet collaborators include Microsoft AI for Earth headed by Lucas Joppa, AMS Committee on Al applications for Environmental Science, ClimateNet at Berkeley Lab, and Stanford ML Group for Climate Change. He received his PhD as a joint Commonwealth scholar with CSIRO at University of New South Wales while being a visiting/affiliate researcher at the Harvard-Smithsonian Center for Astrophysics and Research Fellow in the NY Times bestseller on climate solutions, Drawdown. He founded a startup that was a super-session finalist in the Creative Destruction Lab AI-stream.