I am currently a third year CS PhD student at University of Southern California in the CS Theory Group and ODDS Group where I am advised by Meisam Razaviyayn and Vatsal Sharan. I am broadly interested in working on fundamental problems in Optimization and Differential Privacy. Before joining USC, I recieved a B.Tech in Computer Science and Artificial Intelligence from IIIT Delhi. Email: guptadev (at) usc.edu [Google Scholar]

Publications

• Efficient DP-SGD for LLMs with Randomized Clipping
  Enayat Ullah, Sai Aparna Aketi, Devansh Gupta, Huanyu Zhang, Meisam Razaviyayn
  ICML 2026
  TPDP 2026
  

• Memory-Efficient Differentially Private Training with Gradient Random Projection
  Alex Mulrooney, Devansh Gupta, James Flemings, Huanyu Zhang, Murali Annavaram, Meisam Razaviyayn, Xinwei Zhang
  TMLR 2026
  

• Inherent Privacy of Zeroth-Order Projected Gradient Descent
  Devansh Gupta, Meisam Razaviyayn, Vatsal Sharan
  AISTATS 2025
  OPT 2024 at NeurIPS (Oral)
  

• A Stochastic Optimization Framework for Private and Fair Learning From Decentralized Data
  Devansh Gupta, A.S. Poornash, Andrew Lowy, Meisam Razaviyayn
  PPAI 2025 at AAAI
  

• Stochastic Differentially Private and Fair Learning
  Andrew Lowy, Devansh Gupta, Meisam Razaviyayn
  ICLR 2023
  TPDP 2023; AFCP 2022 at NeurIPS (Oral)