Cooperative AI

Stratospheric Aerosol Injection as a Deep Reinforcement Learning Problem
We suggest treating Stratospheric Aersol Injection (SAI) as a high-dimensional control problem, with policies trained according to a context-sensitive reward function within the Deep Reinforcement Learning (DRL) paradigm.
Christian Schroeder de Witt, Thomas Hornigold
PDF Cite
Stratospheric Aerosol Injection as a Deep Reinforcement Learning Problem
Hijacking Malaria Simulators with Probabilistic Programming
We introduce an approach that allows one to treat a large class of population-based epidemiology simulators as probabilistic generative models.
Bradley Gram-Hansen, Christian Schroeder de Witt, Tom Rainforth, Philip HS Torr, Yee Whye Teh, Atılım Güneş Baydin
PDF Cite
Hijacking Malaria Simulators with Probabilistic Programming
The Starcraft Multi-Agent Challenge
We propose the StarCraft Multi-Agent Challenge (SMAC) to measure real progress in MARL.
Mikayel Samvelyan, Tabish Rashid, Christian Schroeder de Witt, Gregory Farquhar, Nantas Nardelli, Tim GJ Rudner, Chia-Man Hung, Philip HS Torr, Jakob Foerster, Shimon Whiteson
PDF Cite
The Starcraft Multi-Agent Challenge
Biological Evolution and Genetic Algorithms: Exploring the Space of Abstract Tile Self-Assembly
A physically-motivated genetic algorithm (GA) and full enumeration for a tile-based model of self-assembly (JaTAM) is implemented using a graphics processing unit (GPU).
Christian Schroeder de Witt
PDF Cite
Biological Evolution and Genetic Algorithms: Exploring the Space of Abstract Tile Self-Assembly