Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2505.17323 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866913058581905408 |
|---|---|
| author | Mon-Williams, Ruaridh Taylor-Davies, Max Mieczkowski, Elizabeth Velez, Natalia Bramley, Neil R. Wang, Yanwei Griffiths, Thomas L. Lucas, Christopher G. |
| author_facet | Mon-Williams, Ruaridh Taylor-Davies, Max Mieczkowski, Elizabeth Velez, Natalia Bramley, Neil R. Wang, Yanwei Griffiths, Thomas L. Lucas, Christopher G. |
| contents | Humans are remarkably adept at collaboration, able to infer the strengths and weaknesses of new partners in order to work successfully towards shared goals. To build AI systems with this capability, we must first understand its building blocks: does such flexibility require explicit, dedicated mechanisms for modelling others -- or can it emerge spontaneously from the pressures of open-ended cooperative interaction? To investigate this question, we train simple model-free RNN agents to collaborate with a population of diverse partners. Using the `Overcooked-AI' environment, we collect data from thousands of collaborative teams, and analyse agents' internal hidden states. Despite a lack of additional architectural features, inductive biases, or auxiliary objectives, the agents nevertheless develop structured internal representations of their partners' task abilities, enabling rapid adaptation and generalisation to novel collaborators. We investigated these internal models through probing techniques, and large-scale behavioural analysis. Notably, we find that structured partner modelling emerges when agents can influence partner behaviour by controlling task allocation. Our results show that partner modelling can arise spontaneously in model-free agents -- but only under environmental conditions that impose the right kind of social pressure. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_17323 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Partner Modelling Emerges in Recurrent Agents (But Only When It Matters) Mon-Williams, Ruaridh Taylor-Davies, Max Mieczkowski, Elizabeth Velez, Natalia Bramley, Neil R. Wang, Yanwei Griffiths, Thomas L. Lucas, Christopher G. Artificial Intelligence Machine Learning Humans are remarkably adept at collaboration, able to infer the strengths and weaknesses of new partners in order to work successfully towards shared goals. To build AI systems with this capability, we must first understand its building blocks: does such flexibility require explicit, dedicated mechanisms for modelling others -- or can it emerge spontaneously from the pressures of open-ended cooperative interaction? To investigate this question, we train simple model-free RNN agents to collaborate with a population of diverse partners. Using the `Overcooked-AI' environment, we collect data from thousands of collaborative teams, and analyse agents' internal hidden states. Despite a lack of additional architectural features, inductive biases, or auxiliary objectives, the agents nevertheless develop structured internal representations of their partners' task abilities, enabling rapid adaptation and generalisation to novel collaborators. We investigated these internal models through probing techniques, and large-scale behavioural analysis. Notably, we find that structured partner modelling emerges when agents can influence partner behaviour by controlling task allocation. Our results show that partner modelling can arise spontaneously in model-free agents -- but only under environmental conditions that impose the right kind of social pressure. |
| title | Partner Modelling Emerges in Recurrent Agents (But Only When It Matters) |
| topic | Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2505.17323 |