Open-Ended Learning Leads to Generally Capable Agents | Embodied AI Lecture Series at AI2 @allenai
Open-Ended Learning Leads to Generally Capable Agents | Embodied AI Lecture Series at AI2  @allenai
Uploaded October 2021 | Updated November 2026, 3 days ago
Embodied AI Lecture Series @ PRIOR
For previous recordings and upcoming talks visit prior.allenai.org/lectures

Open-Ended Learning Leads to Generally Capable Agents
Max Jaderberg • DeepMind • (10/1/21)

Abstract:
In this talk I will cover our recent publication "Open-Ended Learning Leads to Generally Capable Agents" (deepmind.com/blog/article/generally-capable-agents-emerge-from-open-ended-play). 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 open-ended 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 held-out test tasks including hide-and-seek and capture-the-flag, and we will explore these results and the emergent behaviours and representations of the agent.
Open-Ended Learning Leads to Generally Capable Agents | Embodied AI Lecture Series at AI2From LLMs to Agents: Generalizability from the Inside OutData-Centric Approaches to Adapting Foundation ModelsThe University of Washington eScience Institute: a Home for Data-Intensive DiscoveryGeneralization for Robot Learning In The Wild | Embodied AI Lecture series at AI2Towards Generalist Agents for Accelerating Scientific Discovery171Modular Language ModelsTowards robust long-form text generation systemsDigital Socrates: Evaluating LLMs through Explanation CritiquesAi2: The best AI tools arent built for scientists. Theyre built with them.Ai2s Noah Smith on the next Olmo modelBeyond End-to-end: Decomposed Modeling and Representations in NLP
Ai2 |

Open-Ended Learning Leads to Generally Capable Agents | Embodied AI Lecture Series at AI2

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER