I am a Research Scientist at Google DeepMind in Mountain View. My research focuses on machine learning models that can accrue knowledge over time, becoming more sample and compute efficient. To this end, I study continual learning, transfer learning, meta-learning, and multitask learning.

I completed my PhD in Machine Learning at the University of Oxford as a Rhodes Scholar, advised by Prof. Philip H. S. Torr. My thesis focused on continual learning for efficient machine learning. Prior to that, I studied Electrical Engineering at the University of Engineering and Technology (UET), Lahore.

Publications

For a complete and up-to-date list of my publications, please visit my Google Scholar profile.

Experience

• Research Scientist Apr 2021 - Present
  Google DeepMind, Mountain View
  Frontier AI Unit. Working on continual learning and foundation models.
• Software Developer Jun 2013 - Aug 2016
  Mentor Graphics, Lahore
  Worked on the design and development of hypervisors for embedded systems.

Teaching

• Machine Learning Tutorials 2018, 2020
  Stanford House, University of Oxford
• Networking Tutorials 2018, 2019
  University of Oxford
• Operating Systems Tutorials 2018, 2019
  University of Oxford

Service

Conference Reviewer: ICML, NeurIPS, ICLR, CVPR, ICCV, ECCV, CoLLAs, AISTATS.

Journal Reviewer: IEEE TPAMI, TMLR, IEEE Transaction on Multimedia.