About

I am an applied research scientist in the GTAR team at JPMorganChase, where I work on randomized methods with applications in quantum-inspired and machine learning algorithms.

Previously, I was a postdoctoral research scholar in the department of Medical Physics at the Sloan Kettering Institute hosted by Masoud Zarepisheh. I graduated with a PhD from the Manning College of Information and Computer Sciences at the University of Massachusetts, Amherst, where I was advised by the wonderful Cameron Musco and was a member of the Theory Group.

My work focuses on approximation of matrix methods using fast linear algebra with applications to real world problems. Broadly, my work can be seen as a set of tools that can be utilized to accomplish various tasks within machine learning, in particular neural networks.

Prior to this I worked as Visiting Researcher at the Electronics and Communication Sciences Unit, Indian Statistical Institute, Kolkata with Professor Dipti Prasad Mukherjee. I completed my Masters in Computer Science in the year 2015, from the same institute. My primary research interest was on non-linear learning models for pre-image computation of image datasets.

In my leisure I enjoy playing guitar, badminton and birding. I am an ardent fan of absurdist fiction, philosophical fiction, epic fictions, and graphic novels. I love Seinfeld and Curb Your Enthusiasm and consider them the pinnacle of observational and conversational comedy.

News

  • (June 2026) Our paper, Adaptive and Robust Watermark for Generative Tabular Data was accepted for publication at UAI 2026.
  • (May 2026) Our paper, MetaTT: A Global Tensor-Train Adapter for Parameter-Efficient Fine-Tuning was accepted for publication at TMLR 2026.
  • (May 2026) Presented our work, GPU-Parallelizable Randomized Sketch-and-Precondition for Linear Regression using Sparse Sign Sketches at PDSEC 2026.
  • (April 2026) Our paper, GPU-Parallelizable Randomized Sketch-and-Precondition for Linear Regression using Sparse Sign Sketches won the best paper award at PDSEC 2026.
  • (September 2025) Our paper, A Unified Framework for Provably Efficient Algorithms to Estimate Shapley Values was accepted for publication at Neurips 2025.
  • (October 2024) Our paper, Improved Spectral Density Estimation via Explicit and Implicit Deflation was accepted for publication at SODA 2025.