About Me
I'm a researcher in the field of artificial intelligence. Specifically, I work on creating agents that have to perceive their world and choose actions to achieve their goals (often in the presence of teammates or adversaries), putting me in the areas of reinforcement learning and computational game theory. Most of my work involves multi-agent settings, where humans and AI agents interact in a shared environment. Those environments are often games: sometimes abstract games such as poker or board games, sometimes video games with more depth and complexity, and sometimes high-fidelity virtual reality games meant to emulate our reality.
I'm currently advising startups and doing research in the Agentic LLM space. My last few positions have been:
- Co-founder of Artificial.Agency, where we applied LLM-based agents to video games to drive NPC behavior and game master systems, hoping to bring the style of open-ended creative gameplay found in tabletop RPGs to the video game setting.
- Senior Research Scientist at Google DeepMind, where I worked on reinforcement learning in rich and continual multi-agent / multi-human environments.
- PhD and MSc under Michael Bowling at the University of Alberta, where I was part of the Computer Poker Research Group. We took poker-playing AI from amateur level in 2006 to beating humans in heads-up limit hold'em in 2008, and then solved heads-up limit in 2015 and beat human pros in heads-up no-limit in 2017.
News
April 2023: Co-founded Artificial.Agency
Three of my DeepMind colleagues and I have left the company to start a startup: Artificial.Agency! We're applying RL and LLM techniques to video games, focusing not on chat or dialogue but on the behavior of characters and game systems.
May 2022: Multi-agent Economics
My most recent paper, on multi-agent economics, just hit arXiv: [Local link]. I've been working on this since around January 2018, shortly after joining DeepMind, and after four years I'm excited to share it.
July 2017: DeepMind
I'm now a Research Scientist at DeepMind's new Edmonton office.
March 2017: DeepStack
Our latest Science paper was just released, on DeepStack: game-changing algorithm for tackling large imperfect information games. In December 2016, it became the first program to beat human pros at heads-up no-limit Texas hold'em. [DeepStack.ai]
September 2016: Joined Cogitai
I'm now working at Cogitai.
January 2016: PhD defended!
I defended my PhD this afternoon (January 14th)! Details and thesis are here: [HTML].