mlsanityData to 13 Oct 2026

Self-Explaining Deviations for Coordination

arXiv 2207.12322 ↗First posted 13 Jul 2022cs.AI cs.LGcited by 1

Not on the map. Why?

Abstract (arXiv)

Fully cooperative, partially observable multi-agent problems are ubiquitous in the real world. In this paper, we focus on a specific subclass of coordination problems in which humans are able to discover self-explaining deviations (SEDs). SEDs are actions that deviate from the common understanding of what reasonable behavior would be in normal circumstances. They are taken with the intention of causing another agent or other agents to realize, using theory of mind, that the circumstance must be abnormal. We first motivate SED with a real world example and formalize its definition. Next, we introduce a novel algorithm, improvement maximizing self-explaining deviations (IMPROVISED), to perform SEDs. Lastly, we evaluate IMPROVISED both in an illustrative toy setting and the popular benchmark setting Hanabi, where it is the first method to produce so called finesse plays, which are regarded as one of the more iconic examples of human theory of mind.

Key facts. Cited by 1 paper in the map (data through 13 Oct 2026). Builds on 16 of its 35 references that are in the map. First posted on arXiv 13 Jul 2022.

Builds on

16 of 35 references.

References · 16

  1. 2018QMIX: Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learningcited by 700
  2. 2019The StarCraft Multi-Agent Challengecited by 592
  3. 2017DeepStack: Expert-Level Artificial Intelligence in No-Limit Pokercited by 228
  4. 2019The Hanabi Challenge: A New Frontier for AI Researchcited by 194
  5. 2020"Other-Play" for Zero-Shot Coordinationcited by 194
  6. 2021Collaborating with Humans without Human Datacited by 140
  7. 2020Combining Deep Reinforcement Learning and Search for Imperfect-Information Gamescited by 110
  8. 2021Off-Belief Learningcited by 73
  9. 2018Bayesian Action Decoder for Deep Multi-Agent Reinforcement Learningcited by 68
  10. 2019Simplified Action Decoder for Deep Multi-Agent Reinforcement Learningcited by 48
  11. 2019Rethinking Formal Models of Partially Observable Multiagent Decision Makingcited by 35
  12. 2019Improving Policies via Search in Cooperative Partially Observable Gamescited by 33
4 more
  1. 2018Learning in POMDPs with Monte Carlo Tree Searchcited by 15
  2. 2021Solving Common-Payoff Games with Approximate Policy Iterationcited by 15
  3. 2020On Bellman's Optimality Principle for zs-POSGscited by 2
  4. 2021Common Information Belief based Dynamic Programs for Stochastic Zero-sum Games with Competing Teamscited by 1

Built on by

Cited by year

Citing papers per year: 2023: 1.

1 paper.

  1. 2023JaxMARL: Multi-Agent RL Environments and Algorithms in JAXcited by 61