Bisimulations for Knowing How Logics

Fervari, Raul; Velázquez–Quesada, Fernando R.; Wang, Yanjing

doi:10.1017/s1755020321000101

Cited by 9 publications

(2 citation statements)

References 45 publications

(82 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The representations for observations, actions, or underlying messages are widely studied in MARL. Some works [3,8] use bisimulation metrics to extract the latent embeddings from observations. [19,1,31,30] attempt to learn action representations to assist multi-agent policy learning.…”

Section: Related Workmentioning

confidence: 99%

Learning to Collaborate by Grouping: A Consensus-Oriented Strategy for Multi-Agent Reinforcement Learning

Ruan,

Hao,

et al. 2023

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

Multi-agent systems require effective coordination between groups and individuals to achieve common goals. However, current multi-agent reinforcement learning (MARL) methods primarily focus on improving individual policies and do not adequately address group-level policies, which leads to weak cooperation. To address this issue, we propose a novel Consensus-oriented Strategy (CoS) that emphasizes group and individual policies simultaneously. Specifically, CoS comprises two main components: (a) the vector quantized group consensus module, which extracts discrete latent embeddings that represent the stable and discriminative group consensus, and (b) the group consensus-oriented strategy, which integrates the group policy using a hypernet and the individual policies using the group consensus, thereby promoting coordination at both the group and individual levels. Through empirical experiments on cooperative navigation tasks with both discrete and continuous spaces, as well as google research football, we demonstrate that CoS outperforms state-of-the-art MARL algorithms and achieves better collaboration, thus providing a promising solution for achieving effective coordination in multi-agent systems.

show abstract

Section: Related Workmentioning

confidence: 99%

Learning to Collaborate by Grouping: A Consensus-Oriented Strategy for Multi-Agent Reinforcement Learning

Ruan,

Hao,

et al. 2023

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

show abstract

“…Of course, for the Patriot to know that it has a strategy is not the same as to know what the strategy is. The distinction between "knowing that a strategy exists" and "knowing what the strategy is", in the case of the traditional (indistinguishability-relation-based) knowledge, has been studied in various logics of know-how strategies ( Ågotnes and Alechina 2019;Naumov and Tao 2017;Fervari et al 2017;Naumov and Tao 2018c,b,a;Fervari, Velázquez-Quesada, and Wang 2021;Li and Wang 2021).…”

Section: Introductionmentioning

confidence: 99%

The Logic of Doxastic Strategies

Jiang,

Naumov

2024

AAAI

View full text Add to dashboard Cite

In many real-world situations, there is often not enough information to know that a certain strategy will succeed in achieving the goal, but there is a good reason to believe that it will. The paper introduces the term "doxastic" for such strategies. The main technical contribution is a sound and complete logical system that describes the interplay between doxastic strategy and belief modalities.

show abstract