Nishanth Shetty

PhD Student at the Indian Institute of Science.

prof_pic.jpg

Spectrum Lab, C-239

EE, IISc, Bangalore-560012

I am Nishanth, a PhD student in the Department of Electrical Engineering, Indian Institute of Science, Bengaluru, supervised by Prof. Chandra Sekhar Seelamantula. I study the score function, the gradient of the log-density (Stein score), as a unifying object across sampling (generative modelling), denoising (signal processing), and parameter estimation in unnormalised models (statistics), developing theory and algorithms for inverse problems in imaging and signal estimation. My PhD is funded by the Prime Minister’s Research Fellowship (PMRF) and the Qualcomm Innovation Fellowship 2023, 2024 and 2026.

Before IISc, I did my B.Tech in Electronics and Communication Engineering at PES University, Bengaluru.

news

May 14, 2026 Recognized as a Gold Reviewer for ICML 2026! :tada:
May 13, 2026 Received the outstanding teaching assistant award from the Department of Electrical Engineering! :tada:
Oct 3, 2025 New preprint out on arxiv - Dale meets Langevin: A Multiplicative Denoising Diffusion Model! :tada:
Dec 21, 2024 Monte Carlo Score Matching for Image Generation got accepted at ICASSP 2025, Hyderabad! :tada:
Jul 24, 2024 We won the Qualcomm Innovation Fellowship 2024 for our proposal titled Exploring the Connections Between Brownian Motion and Diffusion Models! :tada:

selected publications

  1. Monte Carlo Score Matching for Image Generation
    Nishanth Shetty, and Chandra Sekhar Seelamantula
    In Proc. IEEE Int. Conf. Acoust., Speech, Signal Process. (ICASSP), 2025
  2. MILD: Momentum Imbued Langevin Dynamics for Faster Sampling
    Nishanth Shetty, Manikanta Bandla, Nishit Neema, Siddarth Asokan, and Chandra Sekhar Seelamantula
    In Proc. IEEE Int. Conf. Acoust., Speech, Signal Process. (ICASSP), 2024
  3. f-GANs Settle Scores!
    Siddarth Asokan, Nishanth Shetty, Aadithya Srikanth, and Chandra Sekhar Seelamantula
    In NeurIPS 2023 Workshop on Diffusion Models, 2023