About me

I am a third year PhD student at the University of Alberta, working in Reinforcement Learning. I am fortunate to be advised by Matthew E. Taylor at the UofA. I also spend my time at MSR Montreal, where I work with Philip Bachman. Currently, I am visiting UC Berkeley, where I am advised by Sergey Levine.

Previously, I was an AI resident at Facebook AI Research, Menlo Park, where I got the chance to work with Mohammad Ghavamzadeh. Even before, I was an undergraduate at IIT Madras, where I got introduced to RL research by Balaraman Ravindran.

Currently, I am working on learning efficient representations, particularly for offline video datasets. I am also interested in combining backprop with local update mechanisms for training large neural networks.

Contact me at my first name dot last name at gmail dot com

Research

Ignorance is Bliss: Robust Control via Information Gating. Manan Tomar, Riashat Islam, Sergey Levine, Philip Bachman. Pre-print.

Agent-Controller Representations: Principled Offline RL with Rich Exogenous Information. Riashat Islam*, Manan Tomar*, Alex Lamb, Yonathan Efroni, Hongyu Zang, Aniket Didolkar, Dipendra Misra, Xin Li, Harm van Seijen, Remi Tachet des Combes, John Langford. Offline RL Workshop at NeurIPS 2022.

Representation Learning in Deep RL via Discrete Information Bottleneck. Riashat Islam, Hongyu Zang, Manan Tomar, Aniket Didolkar, Md Mofijul Islam, Samin Yeasar Arnob, Tariq Iqbal, Xin Li, Anirudh Goyal, Nicolas Heess, Alex Lamb. AISTATS 2023.

Learning Minimal Representations with Model Invariance. Manan Tomar, Amy Zhang, Matthew E. Taylor. Pre-print.

Learning Representations for Pixel-based Control: What Matters and Why?. Manan Tomar*, Utkarsh A. Mishra*, Amy Zhang, Matthew E. Taylor. TMLR 2022.

Model-Invariant State Abstractions for Model-based Reinforcement Learning. Manan Tomar, Amy Zhang, Roberto Calandra, Matthew E. Taylor, Joelle Pineau. Pre-print; Spotlight at Sparsity in Neural Networks workshop.

Mirror Descent Policy Optimization. Manan Tomar, Lior Shani, Yonathan Efroni, Mohammad Ghavamzadeh. Contributed talk at the DeepRL NeurIPS 2020 workshop; ICLR 2022.

Multi-step Greedy Reinforcement Learning Algorithms. Manan Tomar*, Yonathan Efroni*, Mohammad Ghavamzadeh. ICML 2020.

Successor Options : An Option Discovery Algorithm for Reinforcement Learning. Rahul Ramesh*, Manan Tomar*, Balaraman Ravindran. IJCAI 2019.

MaMiC: Macro and Micro Curriculum for Robotic Reinforcement Learning. Manan Tomar, Akhil Sathuluri, Balaraman Ravindran. AAMAS 2019.