Center for Human-Compatible AI

Web application where humans can play Overcooked with AI agents.

HumanCompatibleAI

A benchmark environment for fully cooperative human-AI performance.

HumanCompatibleAI

(Experimental) Inverse reinforcement learning from trajectories generated by multiple agents with different (but correlated) rewards

HumanCompatibleAI

Reward Learning by Simulating the Past

HumanCompatibleAI

Benchmark environments for reward modelling and imitation learning algorithms.

HumanCompatibleAI

A prompt injection game to collect data for robust ML research

HumanCompatibleAI

Reproduction of OpenAI and DeepMind's "Deep Reinforcement Learning from Human Preferences"

HumanCompatibleAI

Testing ranking algorithms to improve social cohesion

HumanCompatibleAI

Dataset for the Tensor Trust project

HumanCompatibleAI