Open Ended Learning Leads to Generally Capable Agents, AI2
Embodied AI Lecture Series PRIOR For previous recordings and upcoming talks visit OpenEnded Learning Leads to Generally Capable Agents Max Jaderberg DeepMind (10, 1, 21) Abstract: In this talk I will cover our recent publication OpenEnded Learning Leads to Generally Capable Agents In this work we turn our attention to how to create embodied agents in simulation that can generalise to unseen test tasks and exhibit generally capable behaviour. I will introduce our XLand procedurally generated environment, and the openended learning algorithms that allow us to train agents to cover this vast environment space. This results in agents that are capable across a wide range of heldout test tasks including hideandseek and capturetheflag, and we will explore these results and the emergent behaviours and representations of the agent.
|
|