I am a Computer Scientist with the Data Analysis Group at the Lawrence Livermore National Laboratory, a Federally Funded Research Institute. My research interests broadly include computer vision, machine learning, and high dimensional data analysis. I enjoy collaborating with scientists and researchers across several disciplines of science, and engineering. I also actively review for Vision and ML conferences, notably — CVPR, ICCV, AAAI, ICML, ECCV.
Contact: undefined

My resume (pdf, updated Jan 2020).


  • (Feb 2020) MimicGAN has been accepted for publication in IJCV’s special issue on GANs! [arXiv]
  • (Feb 2020) Chairing a special session on Generative modeling for images & videos at Asilomar 2020.
  • (Jan 2020) Organizing a mini-symposium at SIAM Mathematics of Data Science on machine learning under different kinds of constraints in May 2020. We have four exciting talks planned!
  • (Jan 2020) Rate Invariant Autoencoding of Time-Series led by Kaushik and Suhas will appear at ICASSP 2020. [preprint]
  • (Dec 2019) 2 new preprints are available:
    • MimicGAN: Robust Projection onto Image Manifolds with Corruption Mimicking [arXiv]
      Where we make PGD style optimization more robust for encoding images accurately under unknown distribution shifts.
    • Improved Surrogates in Inertial Confinement Fusion with Manifold and Cycle Consistencies [arXiv]
      Where we use CycleGAN style training for surrogate modeling in ICF with many favourable properties.
  • (Dec 2019) Short papers presented at NeurIPS 2019 workshops:
    • Improving Limited Angle CT Reconstruction with a Robust GAN Prior
      Deep-Inverse Workshop [paper]
    • Extreme Few-view CT Reconstruction using Deep Inference
      Deep-Inverse Workshop [paper]
    • Exploring Physical Generative Models with Scientific Priors
      ML for Physical Sciences Workshop [paper]
    • Designing Deep Inverse Models for History Matching in Reservoir Simulations
      ML for Physical Sciences Workshop [paper]
    • Modeling Human Brain Connectomes using Structured Neural Networks
      Graph Representation Learning Workshop [paper]