During the day I'm a Research Engineer at DeepMind where I work on challenging open problems in machine learning and artificial intelligence. My current focus sits at the intersection of Reinforcement Learning, Multi-Agent, Game Theory and Continuous Control.
I grew up in China, moved to France in 2011 to study and I'm now living and working in London.
Emergent Coordination though Competition
Siqi Liu, Guy Lever, Josh Merel, Saran Tunyasuvunakool, Nicolas Heess, Thore GraepelICLR 2019 | paper | website | code
Hierarchical Visuomotor Control of Humanoids
Josh Merel, Arun Ahuja, Vu Pham, Saran Tunyasuvunakool, Siqi Liu, Dhruva Tirumala, Nicolas Heess, Greg WayneICLR 2019 | paper | demo
Reinforcement Learning Agents acquire Flocking and Symbiotic Behaviour in Simulated Ecosystems
Peter Sunehag, Guy Lever, Siqi Liu, Josh Merel, Nicolas Heess, Joel Z Leibo, Edward Hughes, Tom Eccles, Thore GraepelALIFE 2019 | paper
The Body is not a Given: Joint Agent Policy Learning and Morphology Evolution
Dylan Banarse, Yoram Bachrach, Siqi Liu, Chrisantha Fernando, Nicolas Heess, Pushmeet Kohli, Guy Lever, Thore GraepelAAMAS 2019 | paper
Observational Learning by Reinforcement Learning
Diana Borsa, Nicolas Heess, Bilal Piot, Siqi Liu, Leonard Hasenclever, Remi Munos, Olivier PietquinAAMAS 2019 | paper
Busy Person Patterns | link
James F. Kile, Donald J. Little, Samir Shah
Patterns that help you make the most out of your time.
Laws of Tech: Commoditize Your Complement | link
Interesting take: commoditizing your complement leads to lasting monopoly in tech.
Baby-Sitting the Economy | link
A simple analogy for economic cycles for non-economist.