Aniket Das

About Me

I am a Pre-Doctoral Researcher in the Machine Learning and Optimization (MLO) Team at Google AI, Bangalore. Prior to this, I was an undergraduate at Indian Institute of Technology Kanpur (IITK) double majoring in Electrical Engineering and Mathematics. During my time at IITK and Google, I have been fortunate to learn from an amazing set of mentors (in no particular order) : Praneeth Netrapalli (Google), Dheeraj Nagaraj (Google), Arun Suggala (Google), Michael Muehlebach (MPI-IS), Bernhard Schölkopf (MPI-IS) and Sandeep Juneja (TIFR).

I am applying to PhD positions for the Fall 2024 cycle. Here is a link to my CV

Interests

My interests lie at the intersection of Applied Probability, Statistics and Theoretical Computer Science. Specific interests include:

• Sampling and Markov Chains : Sampling from High-Dimensional Densities, Spin Systems, Wasserstein Gradient Flows, Bakry-Emery Theory

• Applied Probability : Concentration of Measure, Random Matrix Theory, Optimal Transport, High-Dimensional Statistics

• Continuous Optimization : Nonconvex Optimization, Minimax Optimization, Stochastic Optimization

News

• [Nov. 2023] Gave a talk at the MSR-IISc Theory Seminar on analyzing sampling through the lens of optimization.
• [Oct. 2023] Our paper on Fast Finite-Particle Convergence of Stein Variational Gradient Descent is accepted for an Oral Presentation at OTML Workshop, NeurIPS 2023
• [Sept. 2023] Our paper on Fast Finite-Particle Convergence of Stein Variational Gradient Descent is accepted as a Spotlight Paper at NeurIPS 2023
• [May 2023] Our papers on Stochastic Gradient Langevin Dynamics and Near Optimal Heteroscedastic Regression are accepted at COLT 2023.
• [Sept. 2022] Our paper on Sampling Without Replacement for Finite-sum Minimax Optimization is accepted at NeurIPS 2022
• [July 2022] Joined the MLO Team at Google AI, Bangalore as a Pre-Doctoral Researcher
• [June 2022] Graduated from IIT Kanpur with a Double Major in EE and Math.

Selected Publications

• Provably Fast Finite-Particle Variants of SVGD via Virtual Particle Stochastic Approximation
  with Dheeraj Nagaraj
  Spotlight at NeurIPS 2023 [Paper]
  Oral Presentation at OTML Workshop, NeurIPS 2023

• Utilising the CLT Structure in Stochastic Gradient based Sampling : Improved Analysis and Faster Algorithms
  with Dheeraj Nagaraj and Anant Raj
  COLT 2023 [Paper]

• Near Optimal Heteroscedastic Regression with Symbiotic Learning
  with Dheeraj Baby, Dheeraj Nagaraj and Praneeth Netrapalli
  COLT 2023 [Paper]

• Sampling without Replacement Leads to Faster Rates in Finite-Sum Minimax Optimization
  with Bernhard Schölkopf and Michael Muehlebach
  NeurIPS 2022 [Paper]

Service

• Reviewer : JMLR, NeurIPS, AISTATS
• Co-ordinator : Special Interest Group in Machine Learning, IIT Kanpur
• Undergrad Mentor : Programming Club, IIT Kanpur and Stamatics (Math Club), IIT Kanpur

Fun

I like to cook, talk about videogame lore, Evangelion, Kafka and Cioran.

Powered by Jekyll and Minimal Light theme.