Too Many Cooks: Bayesian Inference for Coordinating Multi‐Agent Collaboration
Open accessThe DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index.
Abstract
Collaboration requires agents to coordinate their behavior on the fly, sometimes cooperating to solve a single task together and other times dividing it up into sub‐tasks to work on in parallel. Underlying the human ability to collaborate is theory‐of‐mind (ToM), the ability to infer the hidden mental states that drive others to act. Here, we develop Bayesian Delegation, a decentralized multi‐agent learning mechanism with these abilities. Bayesian Delegation enables agents to rapidly infer the hidden intentions of others by inverse planning. We test Bayesian Delegation in a suite of multi‐agent Markov decision processes inspired by cooking problems. On these tasks, agents with Bayesian Delegation coordinate both their high‐level plans (e.g., what sub‐task they should work on) and their low‐level actions (e.g., avoiding getting in each other's way). When matched with partners that act using the same algorithm, Bayesian Delegation outperforms alternatives. Bayesian Delegation is also a capable ad hoc collaborator and successfully coordinates with other agent types even in the absence of prior experience. Finally, in a behavioral experiment, we show that Bayesian Delegation makes inferences similar to human observers about the intent of others. Together, these results argue for the centrality of ToM for successful decentralized multi‐agent collaboration.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Playing Atari with Deep Reinforcement Learning
Mnih, V., Kavukcuoglu, K., Silver, D., Graves, A., Antonoglou, I., Wierstra, D., & Riedmiller, M. (2013) · arXiv · doi:10.48550/arxiv.1312.5602 · Cited by 9,471
Watkins, C. J. C. H., & Dayan, P. (1992) · Machine Learning · doi:10.1007/bf00992698 · Cited by 9,083
STRIPS: A New Approach to the Application of Theorem Proving to Problem Solving
Fikes, R. E., & Nilsson, N. J. (1971) · Artificial Intelligence · doi:10.1016/0004-3702(71)90010-5 · Cited by 4,882 · cited 3×
“For instance, for the STRIPS instance of the recipe Tomato , the initial state is the initial configuration of the environment (i.e., all objects and their states), the specification of the goal state is Delivery[Plate[Tomato.chopped]] , and the sub-tasks are the Merge operators in Fig.”
InfluentialUnderstanding and sharing intentions: The origins of cultural cognition
Tomasello, M., Carpenter, M., Call, J., Behne, T., & Moll, H. (2005) · Behavioral and Brain Sciences · doi:10.1017/s0140525x05000129 · Cited by 4,882
Building machines that learn and think like people
Lake, B. M., Ullman, T. D., Tenenbaum, J. B., & Gershman, S. J. (2017) · Behavioral and Brain Sciences · doi:10.1017/s0140525x16001837 · Cited by 2,550
Self-citationBuilding Machines That Learn and Think Like People
Lake, B. M., Ullman, T. D., Tenenbaum, J. B., & Gershman, S. J. (2016) · arXiv · doi:10.48550/arxiv.1604.00289 · Cited by 2,550
Self-citationYoung, H. P. (1993) · Econometrica · doi:10.2307/2951778 · Cited by 2,399
The Grammar of Society: The Nature and Dynamics of Social Norms
Bicchieri, C. (2005) · doi:10.1017/cbo9780511616037 · Cited by 1,991
A Natural History of Human Thinking
Tomasello, M. (2014) · doi:10.4159/9780674726369 · Cited by 1,923
Henrich, J. (2016) · doi:10.1515/9781400873296 · Cited by 1,002
Action understanding as inverse planning
Baker, C. L., Saxe, R., & Tenenbaum, J. B. (2009) · Cognition · doi:10.1016/j.cognition.2009.07.005 · Cited by 920
Self-citationAd Hoc Autonomous Agent Teams: Collaboration without Pre-Coordination
Stone, P., Kaminka, G., Kraus, S., & Rosenschein, J. (2010) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v24i1.7529 · Cited by 323
On the Utility of Learning about Humans for Human-AI Coordination
Carroll, M., Shah, R., Ho, M. K., Griffiths, T. L., Seshia, S. A., Abbeel, P., & Dragan, A. (2019) · arXiv · doi:10.48550/arxiv.1910.05789 · Cited by 168 · cited 3×
“hierarchies of sub-tasks [19, 12, 36]. Most recently, environments inspired by Overcooked have been used in deep reinforcement learning studies where agents are trained using self-play and human data [34, 8]. In contrast, our approach is based on techniques that dynamically learn while interacting rather than requirin…”
BackgroundMethodBounded real-time dynamic programming: RTDP with monotone upper bounds and performance guarantees
McMahan, H. B., Likhachev, M., & Gordon, G. J. (2005) · Proceedings of the 22nd international conference on Machine learning - ICML '05 · doi:10.1145/1102351.1102423 · Cited by 128 · cited 1×
“For details on how BRTDP updates V and Q , see McMahan et al. (2005).”
BackgroundFinding Friend and Foe in Multi-Agent Games
Serrino, J., Kleiman-Weiner, M., Parkes, D. C., & Tenenbaum, J. B. (2019) · arXiv · doi:10.48550/arxiv.1906.02330 · Cited by 33 · cited 1×
“mplexities of coordinated sub-tasks. Finally, there is a growing body of work on multi-agent cooperation in the context of competitive interactions which is complementary to our focus on coordination [17,19,21,22,33]. 2 DECENTRALIZED MARKOV DECISION PROCESSES WITH SUB-TASKS Adecentralizedmulti-agentMarkovdecisionproce…”
BackgroundSelf-citation
Cited by
ProAgent: Building Proactive Cooperative Agents with Large Language Models
Zhang, C., Yang, K., Hu, S., Wang, Z., Li, G., Sun, Y., Zhang, C., Zhang, Z., Liu, A., Zhu, S.-C., et al. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i16.29710 · Cited by 88
“…Strouse et al. 2021; Zhao et al. 2023; Lupu et al. 2021; Lucas and Allen 2022; Li et al. 2023b, 2024) and Theory of Mind (ToM; Hu et al. 2021a; Wu et al. 2021; Wang et al. 2021) to facilitate adaptive policy development for coordinating with various counterparts without prior coordination…”
BackgroundA multifaceted vision of the human-AI collaboration: a comprehensive review
Puerta-Beldarrain, M., Gómez-Carmona, O., Sánchez-Corcuera, R., Casado-Mansilla, D., López-de-Ipiña, D., & Chen, L. (2025) · IEEE Access · doi:10.1109/access.2025.3536095 · Cited by 73
“The Overcooked game has been used for various research purposes, including studying human collaboration across different task loads in human-AI environments [278], [279].”
AGENT: A Benchmark for Core Psychological Reasoning
Shu, T., Bhandwaldar, A., Gan, C., Smith, K. A., Liu, S., Gutfreund, D., Spelke, E., Tenenbaum, J. B., & Ullman, T. D. (2021) · arXiv · doi:10.48550/arxiv.2102.12321 · Cited by 34
Self-citation
Metadata
- Authors
- Wu, S. A., Wang, R. E., Evans, J. A., Tenenbaum, J. B., Parkes, D. C., & Kleiman‐Weiner, M.
- Venue
- Topics in Cognitive Science
- Publisher
- Wiley
- Published
- 2021-04
- Volume
- 13
- Issue
- 2
- Pages
- 414-432
- ISSN
- 1756-8757 (ISSN-L)
- 1756-8765
- OpenAlex
- W3148143011
- Cited by
- 92
- References
- 15
- FWCI
- 23.77
- Type
- journal article
- Language
- en
- Open access
- bronze
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.