About Me


I am a PhD student in Machine Learning supervised by Doina Precup and co-supervised by Joelle Pineau in the Reasoning and Learning Lab at McGill University. The lab is co-affiliated with the Montreal Institute of Learning Algorithms (MILA)

Previously, I was a Masters student at University of Cambridge in the MPhil Machine Learning, Speech and Language Technology program. I completed my Masters degree under the supervision of Zoubin Ghahramani and Yarin Gal at Cambridge Machine Learning Group. My Masters was funded by the Cambridge Commonwealth and International Trust, and I was a member of St John’s College, Cambridge. 

Prior to that I studied Electronic and Electrical Engineering at University College London working under supervision of John Shawe-Taylor and Guy Lever from the Centre of Computational Statistics and Machine Learning and Gatsby Computational and Neuroscience Unit at UCL. I also had a great fortune working with David Silver and collaborated closely with Google DeepMind during my undergraduate thesis.

I was also a summer research student at California Institute of Technology (Caltech) in the Summer Undergraduate Research Fellowship program (SURF) where I worked under supervision of Richard Murray in the Control and Dynamical Systems Lab, under a project in collaboration with the NASA Jet Propulsion Laboratory (JPL)and Keck Institute for Space Studies.  Before that, I was a summer research student in the Machine Learning Group at Johns Hopkins University working under supervision of Suchi Saria.

Research Interests:

My primary research interest is in reinforcement learning, with the goal of developing agents with general purpose artificial intelligence. I am interested in developing knowledge-intensive intelligent systems, and how AI agents can maintain accurate world knowledge in a complex and changing environment. This involves solving how AI agents can build a rich understanding and knowledge representation of the world to justify their understanding. The problem of General Artificial Intelligence is to build intelligent systems that can build layers of abstractions and representations of world knowledge, towards the goal of a continuous learning agent that can maintain its own knowledge and understanding of the world. I believe such milestones can be achieved through research in predictive knowledge and reinforcement learning. I am also broadly interested in applications of deep reinforcement learning and learning driven approaches for robotics and game-playing tasks.

I also have broad interests in applications of deep learning, probabilistic modelling, approximate inference, variational methods and Bayesian approaches to deep learning. I am also interested to work on generative modelling and unsupervised learning which are keys to solving artificial intelligence. Previously, I also worked on data-efficient active learning for image data using uncertainty in deep learning. Prior to that, I also worked on policy gradient methods in reinforcement learning.


January 2017 : Started PhD at the RL Lab / MILA at McGill University.

November 2016 : Paper accepted in NIPS’16 Bayesian Deep Learning workshop. “Active Learning Image Data”

October 2016 : Got married 😀 Starting a new project!

October 2016 : I am teaching a course in Introduction to Reinforcement Learning at both IUB and AIUB in Bangladesh

August 2016 : I submitted my Masters thesis in the MPhil Machine Learning, Speech and Language Technology program

August 2016 : Completed the MPhil Machine Learning, Speech and Language Technology program with an average of 72%

October 2015 : Joined for a Masters in Machine Learning at University of Cambridge

August 2015 : Finished the Caltech SURF program

June 2015 : Finally done with undergraduate studies at UCL

April 2015 : Submitted my undergraduate thesis at UCL