AS

Akash Srivastava

MIT-IBM Watson AI Research Lab, University of Edinburgh

ORCID iD 0000-0002-6218-1770
Follow

About

I am a principal investigator and research scientist at the MIT-IBM AI Research Lab in Cambridge, MA, where I work on self-supervised learning with Prof Pulkit Agarwal’s group, hybrid generative models with Prof Faez Ahmed’s group, density ratio estimation with Prof Michael U. Gutmann and more recently, on machine common-sense and intuitive physics with Prof Josh Tenenbuam’s group and caual discovery with Prof Caroline Uhler. Before joining MIT-IBM, I obtained my PhD at the University of Edinburgh where I worked with Prof Charles Sutton and Prof Michael U. Gutmann on variational inference for generative models and deep learning.

Employment

  • MIT-IBM Watson AI Research Lab Director, Core AI
    2019 - Present

Education

  • University of Edinburgh PhD
    2015 - 2018

Projects & Funding

Projects & funding information is unavailable.

Publications (28)

  • Towards robust and generalizable representations of extracellular data using contrastive learning
    2023 DOI: 10.1101/2023.10.30.564831
    +Save
  • Estimating the Density Ratio between Distributions with High Discrepancy using Multinomial Logistic Regression
    2022
    +Save
  • Simulator-assisted training for interpretable generative models
    2021
    +Save
  • A Bayesian-Symbolic Approach to Reasoning and Learning in Intuitive Physics
    Advances in Neural Information Processing Systems 2021
    +Save
  • Equivariant Contrastive Learning
    arXiv preprint arXiv:2111.00899 2021
    +Save
  • Generative Ratio Matching Networks
    International Conference on Learning Representations 2020
    +Save
  • Improving the Reconstruction of Disentangled Representation Learners via Multi-Stage Modelling
    arXiv preprint arXiv:2010.13187 2020
    +Save
  • Sequential transfer machine learning in networks: Measuring the impact of data and neural net similarity on transferability
    arXiv preprint arXiv:2003.13070 2020
    +Save
  • not-so-BigGAN: Generating High-Fidelity Images on Small Compute with Wavelet-based Super-Resolution
    arXiv preprint arXiv:2009.04433 2020
    +Save
  • not-so-biggan: Generating high-fidelity images on a small compute budget
    arXiv e-prints 2020
    +Save