Pritam Sarkar
Ph.D. Student at Queen's University
Affiliate at Vector Institute

pritam[dot]sarkar[at]queensu[dot]ca

Email Google Scholar LinkedIn GitHub

I will be joining Google Research as a Student Researcher starting from October 2023. I am a PhD candidate at Queen’s University and affiliated with the Vector Institute. My research primarily focuses on self-supervised learning to acquire meaningful representations from unlabelled data, i.e., without (or with minimal) human supervision. Earlier, I interned at Borealis AI in Fall 2022. I joined Queen’s University in 2018 and completed my master’s degree in 2020. Prior to that, I worked as a Software Engineer for a total of approximately 3 years at Infosys and Tech Mahindra. I completed my bachelor’s degree in 2015 from West Bengal University of Technology, India. Other than research, I am passionate about photography and film-making.

Research


I am interested in self/un-supervised learning, computer vision, video understanding, multimodal learning (i.e., vision-language, audio-visual), out-of-distribution generalization, large-scale pretraining of foundational models in vision and multimodal domains, and understanding the high-level context from long videos, among a few others. Please find more about my research here.

News


  • [Sep 23] Our paper got accepted in NeurIPS as a Spotlight.
  • [Aug 23] Accepted an offer from Google to join as a Student Researcher.
  • [Nov 22] 2 papers got accepted in AAAI 2023.
  • [Oct 22] We are organizing AAAI 2023 Workshop on R2HCAI.
  • [Oct 22] Honourable Mention in poster competitions (1.) Robotics and AI Symposium 2022 and (2.) FEAS Research Symposium 2022.
  • [Jun 22] Accepted an offer from Borealis AI for a fall internship as a Machine Learning Research Intern.
  • [Oct 21] Best poster award at Robotics and AI Symposium, Ingenuity Labs, 2021.
  • [Aug 21] We are organizing AAAI 2022 Workshop on HC-SSL.
  • [Aug 21] One paper that I supervised got accepted into HAI 2021.
  • [Mar 21] I received postgraduate affiliation award from Vector Institute.
  • [Dec 20] Our paper CardioGAN got accepted in AAAI 2021.
Click to see more
  • [Aug 20] My first journal/transaction as a first author got accepted in IEEE Trans. of Affective Computing.
  • [Apr 20] Successfully defended my M.A.Sc. thesis. picture
  • [Jan 20] Conference paper on ECG-based SSL got accepted in IEEE ICASSP 2020 for oral presentation.
  • [Sep 19] It was a great experience to visit the University of Cambridge and meet Rosalind Picard. picture
  • [Jun 19] My first paper got accepted for oral presentation in IEEE ACII 2019.
  • [Sep 18] Joined Queen's for master's degree.
  • [Dec 17] Joined Infosys as a Sr. System Engineer.
  • [Nov 15] Joined Tech Mahindra as an Associate Software Engineer.
  • [Jun 15] Completed graduation!

Education


  • PhD at Queen’s University, Canada, 2020 - Present.
  • MASc at Queen’s University, Canada, 2018 - 2020. Link to MASc Thesis.
  • B.Tech at West Bengal University of Technology, India, 2011 - 2015.

Employment


  • Research Assistant at Queen’s University, Kingston, Canada, 2018 - Present.
  • Machine Learning Research Intern at Borealis AI, Toronto, Canada, Fall 2022.
  • Sr. System Engineer at Infosys Ltd., Bangalore, India, 2017 - 2018.
  • Software Engineer at Tech Mahindra Ltd., Hyderabad, India, 2015 - 2017.
  • Teaching Assistant/Guest Lecturer at Queen’s University, Kingston, Canada, 2018 - Present.

Academic Service


Mentorship

  • Vishal Narnaware, Visiting Student at University of Cambridge, co-mentored with Nikhil Churamani, present.
  • Brennan Cruse, PhD at CS, Queen’s University, present.
  • Debaditya Shome, MASc at ECE, Queen’s University, present.
  • Jordan Posen, ECE at Queen’s, undergrad final year project, 2021.
  • Rachel Phinnemore, CS at Queen’s, undergrad final year project, 2020.

Organizing workshops and conferences

Reviewing/PC Member

  • NeurIPS, ICLR, CVPR, ICCV, ECCV, ICASSP, ACII
  • IEEE Transactions on - Affective Computing, Artificial Intelligence

Talks

  • [Jul 2023] at Ingenuity Labs, Queen’s University, Title: Learning withour human supervision
  • [Jan 2023] at Borealis AI, Title: AugESeq: Augmentation improves Event Sequence prediction