Artificial agents have achieved great success in individually challenging simulated environments, mastering the particular tasks they were trained for, with their behaviour even generalising to maps and opponents that were never encountered in training. In this talk we explore our recent work "Open-Ended Learning Leads to Generally Capable Agents" in which we create agents that can perform well beyond a single, individual task, that exhibit much wider generalisation of behaviour to a massive, rich space of challenges. We discuss the design of our environment spanning a vast set of tasks and how open-ended learning processes lead to agents that are generally capable across this space and beyond.

https://arxiv.org/abs/2107.12808