About
I’m a research scientist at Google DeepMind working on world models and general agents. I completed my PhD in Computer Science at the University of Toronto, where I was part of the Machine Learning Group and was co-supervised by Prof. Jimmy Ba and Prof. Sheila McIlraith. During my MSc, I was co-supervised by Prof. Sanja Fidler and Prof. Jimmy Ba. During my undergraduate degree, I studied Engineering Science, majoring in Electrical and Computer Engineering at the University of Toronto. My undergraduate thesis advisor was Prof. Deepa Kundur. I interned at Qualcomm Canada in the Video Processing group as part of my Professional Experience Year for 16 months. Prior to starting my graduate degree, I spent a year working at Intel Programmable Solutions Group (PSG) in the OpenCL Usability team. I have also been a research intern at Borealis AI (with Kevin Luk), Google Brain Toronto (with William Chan and Jamie Kiros), Google Research (with DeLesley Hutchins), Google Brain Robotics (with Ted Xiao), and Google DeepMind (with Vlad Mnih). More details are available in my CV.
I’m also a speedcuber.
Research Interests
My research focuses on building general agents that can learn and act in rich, open-ended environments. I’m particularly interested in world models, multimodal foundation models, and reinforcement learning, including how agents can learn from language, demonstrations, and interaction.
Recent Work
I contributed to Genie 2, a foundation world model for generating playable 3D environments, and SIMA 2, a Gemini-powered generalist agent for virtual worlds. My recent research also includes LMAct, a benchmark for long-context, multimodal in-context imitation learning.
Teaching
I was a Teaching Assistant for the following courses:
- CSC413/2516 Neural Networks and Deep Learning, Winter 2020, Winter 2021 (Head TA), and Winter 2022 (Co-Head TA).
- CSC311 Introduction to Machine Learning, Fall 2019/2020.
- ECE421 Introduction to Machine Learning, Winter 2019. Taught weekly 2-hour tutorials.
- MIE324 Introduction to Machine Intelligence, Fall 2018. Developed 3 new assignments over the summer.
- CSC321 Introduction to Neural Networks, Winter 2018.
- CSC411 Introduction to Machine Learning, Fall 2017.
Talks
- Closing the generalization gap in stochastic optimization through Fisher gradient noise. Vector Institute, Toronto, Canada. February 2018.