Pranshu Malviya

PhD Student at MILA


I recently joined MILA / Polytechnique Montreal as a PhD student supervised by Prof. Sarath Chandar (Chandar Research Lab). My primary area of research is Lifelong Learning.

Before this, I completed my MS in the Dept. of CSE at IIT Madras, supervised by Prof. Balaraman Ravindran and Prof. Sarath Chandar. The title of my MS thesis was “TAG: Task-based Accumulated Gradients for Lifelong learning”. At IITM, I was a part of the Robert Bosch Centre for Data Science and Artificial Intelligence (RBCDSAI) Lab where I worked on several research areas: Lifelong Learning, Causal Inference, Bayesian Networks and Deep Learning. I was also involved in a project in collaboration with Applied Materials. It involved exploring states in the time-series data obtained from the semiconductor manufacturing process.

I completed my undergrad (B.Tech) in Dept. of CSE, IIIT-Bhubaneswar. At IIIT-Bh, I worked on numerous projects at different scales that involved machine learning, image processing, and optimization problems. My Bachelor’s thesis titled “Bayesian Optimization and Deep Learning for parameter tuning in Apache Hadoop”. This project was based on a deep learning approach that leveraged the knowledge of system configuration and test dataset to optimize Hadoop parameters to overcome the limitations of Bayesian optimization.

Link: Resume.