Hello!
I am a postdoctoral research and teaching fellow at the University of British Columbia and the Vector Institute supervised by Prof. Jeff Clune. I am interested in developing autonomous agents that are safe, curious, and can learn in an open-ended manner, particularly driven by recent advances in foundation models together with deep reinforcement learning.
🎉 Introducing The AI Scientist
We recently released The AI Scientist, the first agent that can automate the entire scientific process. Our system can independently form hypotheses, conduct experiments, visualize results, write a paper, and review it. Thank you to my incredible collaborators!
📰 Press Coverage
I have been fortunate to have my work featured in various media outlets, including:
- Science News - Highlighting how AI is being trained to mimic creative human intuition.
- The AI Scientist has been covered in Nature News, VentureBeat, Ars Technica, WIRED, IEEE Spectrum, and Forbes.
- Our UBC lab's work has been covered more generally in Air Street Press, this features: Intelligent Go-Explore, and The AI Scientist!
I previously received my PhD at the University of Oxford supervised by Prof. Michael A. Osborne and Prof. Yee Whye Teh. During the PhD, I was particularly interested in offline reinforcement learning with work in generalization to unseen tasks, uncertainty quantification for offline world models, learning from pixels, and diffusion synthetic data for reinforcement learning. Please do feel free to reach out!
You can find my PhD thesis here.
Recent News
- [9/2024] I am starting as a teaching fellow teaching 3rd year undergraduate and graduate machine learning. See you in CS340!
- [8/2024] We propose Automated Design of Agentic Systems, a new paradigm for automatically designing LLM agents.
- [8/2024] We introduce The AI Scientist, the first fully autonomous agent that can complete the entire scientific process!
- [8/2024] Our work was covered in Science News!
- [5/2024] Excited to introduce Intelligent Go-Explore, empowering foundation model agents to robustly explore in complex environments!
- [5/2024] Check out our new survey on video diffusion models!
- [4/2024] Our work on language-guided control won an outstanding paper award at the GenAI4DM Workshop!
- [1/2024] Excited to begin my postdoc at the University of British Columbia working on open-endedness!
- [11/2023] 2 new papers at NeurIPS workshops, on edge-of-reach problems in offline MBRL and on-policy guided synthetic data generation!
- [10/2023] Excited to rejoin Waymo Research on the Sim Agents team!
- [9/2023] New work: Synthetic Experience Replay - arbtrarily upsampling an agent's experiences accepted at NeurIPS. See you in New Orleans!
- [8/2023] Challenges and opportunities in offline reinforcement learning from visual observations accepted at TMLR.
- [12/2022] Internship project at MSR Cambridge on automated game testing using Go-Explore accepted at IEEE ToG.
- [8/2022] I am interning at Waymo Research supervised by Max Igl.
- [6/2022] Our work on offline RL from pixels won the Best Paper Award at L-DOD.
- [5/2022] Our work on PBT across architectures and hyperparameters was accepted at AutoML.
- [3/2022] I am interning at MSR Cambridge in the Deep RL for Games group.
- [2/2022] Our work on revisiting design choices in offline model-based reinforcement learning received a Spotlight at ICLR!