Open-Ended Learning Leads to Generally Capable Agents

by   Open-Ended Learning Team, et al.

In this work 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 define a universe of tasks within an environment domain and demonstrate the ability to train agents that are generally capable across this vast space and beyond. The environment is natively multi-agent, spanning the continuum of competitive, cooperative, and independent games, which are situated within procedurally generated physical 3D worlds. The resulting space is exceptionally diverse in terms of the challenges posed to agents, and as such, even measuring the learning progress of an agent is an open research problem. We propose an iterative notion of improvement between successive generations of agents, rather than seeking to maximise a singular objective, allowing us to quantify progress despite tasks being incomparable in terms of achievable rewards. We show that through constructing an open-ended learning process, which dynamically changes the training task distributions and training objectives such that the agent never stops learning, we achieve consistent learning of new behaviours. The resulting agent is able to score reward in every one of our humanly solvable evaluation levels, with behaviour generalising to many held-out points in the universe of tasks. Examples of this zero-shot generalisation include good performance on Hide and Seek, Capture the Flag, and Tag. Through analysis and hand-authored probe tasks we characterise the behaviour of our agent, and find interesting emergent heuristic behaviours such as trial-and-error experimentation, simple tool use, option switching, and cooperation. Finally, we demonstrate that the general capabilities of this agent could unlock larger scale transfer of behaviour through cheap finetuning.


page 1

page 2

page 3

page 7

page 17

page 22

page 30

page 41


Heterogeneous Multi-agent Zero-Shot Coordination by Coevolution

Generating agents that can achieve Zero-Shot Coordination (ZSC) with uns...

Learning When to Transfer among Agents: An Efficient Multiagent Transfer Learning Framework

Transfer Learning has shown great potential to enhance the single-agent ...

Emergent Coordination Through Competition

We study the emergence of cooperative behaviors in reinforcement learnin...

AutoDIME: Automatic Design of Interesting Multi-Agent Environments

Designing a distribution of environments in which RL agents can learn in...

MineDojo: Building Open-Ended Embodied Agents with Internet-Scale Knowledge

Autonomous agents have made great strides in specialist domains like Ata...

Evolving Curricula with Regret-Based Environment Design

It remains a significant challenge to train generally capable agents wit...

Enhanced POET: Open-Ended Reinforcement Learning through Unbounded Invention of Learning Challenges and their Solutions

Creating open-ended algorithms, which generate their own never-ending st...