prof_pic_square.jpg

Shubhankar (Shuby) Gahlot

Machine Learning · Research · In harmonic motion between Art & Science

Huntsville, Alabama


I am an oncoming Earth and Ocean Sciences PhD candidate at The University of Victoria, Canada.

Previously, I worked as a research scientist at NASA IMPACT project where I leveraged NASA earth data and machine learning to solve remote sensing problems. As part of the group’s outreach program I also co-organized a joint NASA/IEEE Geoscience and Remote Sensing Society (GRSS) challenge on applying AI to open water flood extent detection. The competition recieved 137 participants from 10 countries and over 400 submissions.

Prior to that I was a Scalable Deep Learning researcher at Oak Ridge National Lab where I worked on developing and benchmarking Deep Learning (machine learning) workflows for the Summit Supercomputer (9,216 IBM Power processors & 27,648 NVDIA V100 GPUs) (fastest in the world, 2019). At ORNL, I also worked on leveraging data optimization techniques for large batch distributed training of deep neural networks.

I graduated with a Master’s in Data Science from Illinois Institute of Technology Chicago and a Bachelor’s in Design from Indian Institute of Technology Guwahati.

In my previous avatar, I was a Human-Computer Interaction designer and product architect. I’ve interests that range from philosophy, science and technology to arts, music and traveling. My current interests include building things out of salvaged materials, training and inference of computer vision and languade models on Edge devices using model quantization, pruning and compression techniques. I am also interested in developing machine learning workflows that can run on the Raspberry Pi cluster that I built. I have always tried to balance my creativity with my scientific pursuit.


news

Mar 16, 2022 Published and open sourced the NASA Flood Extent Data for Machine Learning v1.0 dataset! :satellite: :ocean: :open_file_folder: :tada:
Dec 15, 2021 Presented our work on Leveraging Citizen and Artificial Intelligence for Monitoring and Estimating Hazardous Events at the AGU Fall Meeting 2021 :microphone: :ocean: :umbrella:
Apr 15, 2021 Organized a competition on “Global Flood Detection Challenge”! :earth_americas: :ocean:
Dec 3, 2020 Presented our work at Computational Science and Computational Intelligence (CSCI’20) conference! :microphone:
Nov 2, 2020 Joined NASA IMPACT as a research scientist. :tada:👨🏽‍🔬

latest posts

selected publications

  1. IEEE/ACM DLS
    Strategies to Deploy and Scale Deep Learning on the Summit Supercomputer
    Junqi Yin, Shubhankar Gahlot, Nouamane Laanait, and 4 more authors
    In 2019 IEEE/ACM Third Workshop on Deep Learning on Supercomputers (DLS), Nov 2019
  2. CSCI
    Data optimization for large batch distributed training of deep neural networks
    Shubhankar Gahlot, Junqi Yin, and Mallikarjun Arjun Shankar
    In 2020 International Conference on Computational Science and Computational Intelligence (CSCI), Dec 2020
  3. AGU
    Verb Sense Disambiguation for Densifying Knowledge Graphs in Earth Science
    Ashish Acharya, Carson Davis, Derek Koehl, and 4 more authors
    In AGU Fall Meeting Abstracts, Dec 2021
  4. AGU
    Curating flood extent data and leveraging citizen science for benchmarking machine learning solutions
    Shubhankar Gahlot, Muthukumaran Ramasubramanian, Iksha Gurung, and 3 more authors
    ESS Open Archive eprints, Apr 2022