Autonomous Agents and Multi-Agent Systems最新文献

Augmenting the action space with conventions to improve multi-agent cooperation in Hanabi 用约定扩充动作空间，提高多agent在Hanabi中的合作

IF 2 3区计算机科学

Autonomous Agents and Multi-Agent Systems Pub Date : 2025-05-24 DOI: 10.1007/s10458-025-09709-5

Francois Bredell, Herman A. Engelbrecht, J. C. Schoeman

{"title":"Augmenting the action space with conventions to improve multi-agent cooperation in Hanabi","authors":"Francois Bredell, Herman A. Engelbrecht, J. C. Schoeman","doi":"10.1007/s10458-025-09709-5","DOIUrl":"10.1007/s10458-025-09709-5","url":null,"abstract":"<div><p>The card game <i>Hanabi</i> is considered a strong medium for the testing and development of multi-agent reinforcement learning (MARL) algorithms, due to its cooperative nature, partial observability, limited communication and remarkable complexity. Previous research efforts have explored the capabilities of MARL algorithms within Hanabi, focusing largely on advanced architecture design and algorithmic manipulations to achieve state-of-the-art performance for various number of cooperators. However, this often leads to complex solution strategies with high computational cost and requiring large amounts of training data. For humans to solve the Hanabi game effectively, they require the use of conventions, which often allows for a means to implicitly convey ideas or knowledge based on a predefined, and mutually agreed upon, set of “rules” or principles. Multi-agent problems containing partial observability, especially when limited communication is present, can benefit greatly from the use of implicit knowledge sharing. In this paper, we propose a novel approach to augmenting an agent’s action space using <i>conventions</i>, which act as a sequence of special cooperative actions that span over and include multiple time steps and multiple agents, requiring agents to actively opt in for it to reach fruition. These <i>conventions</i> are based on existing human conventions, and result in a significant improvement on the performance of existing techniques for self-play and cross-play for various number of cooperators within Hanabi.</p></div>","PeriodicalId":55586,"journal":{"name":"Autonomous Agents and Multi-Agent Systems","volume":"39 1","pages":""},"PeriodicalIF":2.0,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10458-025-09709-5.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144125540","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Entropy based blending of policies for multi-agent coexistence 基于熵的多智能体共存策略混合

IF 2 3区计算机科学

Autonomous Agents and Multi-Agent Systems Pub Date : 2025-05-16 DOI: 10.1007/s10458-025-09707-7

David Rother, Franziska Herbert, Fabian Kalter, Dorothea Koert, Joni Pajarinen, Jan Peters, Thomas H. Weisswange

{"title":"Entropy based blending of policies for multi-agent coexistence","authors":"David Rother, Franziska Herbert, Fabian Kalter, Dorothea Koert, Joni Pajarinen, Jan Peters, Thomas H. Weisswange","doi":"10.1007/s10458-025-09707-7","DOIUrl":"10.1007/s10458-025-09707-7","url":null,"abstract":"<div><p>Research on multi-agent interaction involving humans is still in its infancy. Most approaches have focused on environments with collaborative human behavior or a small, defined set of situations. When deploying robots in human-inhabited environments in the future, the diversity of interactions surpasses the capabilities of pre-trained collaboration models. ”Coexistence” environments, characterized by agents with varying or partially aligned objectives, present a unique challenge for robotic collaboration. Traditional reinforcement learning methods fall short in these settings. These approaches lack the flexibility to adapt to changing agent counts or task requirements without undergoing retraining. Moreover, existing models do not adequately support scenarios where robots should exhibit helpful behavior toward others without compromising their primary goals. To tackle this issue, we introduce a novel framework that decomposes interaction and task-solving into separate learning problems and blends the resulting policies at inference time using a goal inference model for task estimation. We create impact-aware agents and linearly scale the cost of training agents with the number of agents and available tasks. To this end, a weighting function blending action distributions for individual interactions with the original task action distribution is proposed. To support our claims we demonstrate that our framework scales in task and agent count across several environments and considers collaboration opportunities when present. The new learning paradigm opens the path to more complex multi-robot, multi-human interactions.</p></div>","PeriodicalId":55586,"journal":{"name":"Autonomous Agents and Multi-Agent Systems","volume":"39 1","pages":""},"PeriodicalIF":2.0,"publicationDate":"2025-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10458-025-09707-7.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144073786","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Distributed Course Allocation with Asymmetric Friendships 基于非对称友谊的分布式课程分配

IF 2 3区计算机科学

Autonomous Agents and Multi-Agent Systems Pub Date : 2025-05-15 DOI: 10.1007/s10458-025-09708-6

Lihi Dery, Tal Grinshpoun, Ilya Khakhiashvili

{"title":"Distributed Course Allocation with Asymmetric Friendships","authors":"Lihi Dery, Tal Grinshpoun, Ilya Khakhiashvili","doi":"10.1007/s10458-025-09708-6","DOIUrl":"10.1007/s10458-025-09708-6","url":null,"abstract":"<div><p>Students’ decisions on whether to take a class are strongly affected by whether their friends plan to take the class with them. A student may prefer to be assigned to a course they like less, just to be with their friends, rather than taking a more preferred class alone. It has been shown that taking classes with friends positively affects academic performance. Thus, academic institutes should prioritize friendship relations when assigning course seats. The introduction of friendship relations results in several non-trivial changes to current course allocation methods. This paper explores how course allocation mechanisms can account for <i>friendships</i> between students and provide a unique, distributed solution. Specifically, we approach the problem by framing it as an asymmetric distributed constraint optimization problem and develop a new dedicated algorithm. Our extensive evaluation includes both simulated data and a study involving 177 students, focusing on their preferences regarding both courses and friendships. The findings indicate that our algorithm achieves significant utility for the students, maintaining fairness in the solution and adhering to the limitations on course seat capacities.</p></div>","PeriodicalId":55586,"journal":{"name":"Autonomous Agents and Multi-Agent Systems","volume":"39 1","pages":""},"PeriodicalIF":2.0,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10458-025-09708-6.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143949595","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Correction: Epistemic selection of costly alternatives: the case of participatory budgeting 修正：昂贵替代方案的认知选择：参与式预算的案例

IF 2 3区计算机科学

Autonomous Agents and Multi-Agent Systems Pub Date : 2025-04-29 DOI: 10.1007/s10458-025-09702-y

Simon Rey, Ulle Endriss

引用次数: 0

Hypertension and total-order forward decomposition optimizations 高血压和全阶正分解优化

IF 2 3区计算机科学

Autonomous Agents and Multi-Agent Systems Pub Date : 2025-04-25 DOI: 10.1007/s10458-025-09705-9

Maurício Cecílio Magnaguagno, Felipe Meneguzzi, Lavindra de Silva

引用次数: 0

Improving execution concurrency in partial-order plans via block-substitution 通过块替换提高部分顺序计划的执行并发性

IF 2 3区计算机科学

Autonomous Agents and Multi-Agent Systems Pub Date : 2025-04-23 DOI: 10.1007/s10458-025-09706-8

Sabah Binte Noor, Fazlul Hasan Siddiqui

{"title":"Improving execution concurrency in partial-order plans via block-substitution","authors":"Sabah Binte Noor, Fazlul Hasan Siddiqui","doi":"10.1007/s10458-025-09706-8","DOIUrl":"10.1007/s10458-025-09706-8","url":null,"abstract":"<div><p>Partial-order plans in AI planning facilitate execution flexibility and several other tasks, such as plan reuse, modification, and decomposition, due to their less constrained nature. A Partial-Order Plan (POP) specifies partial-order over actions, providing the flexibility of executing unordered actions in different sequences. This flexibility can be further extended by enabling parallel execution of actions in the POP to reduce its overall execution time. While extensive studies exist on improving the flexibility of a POP by optimizing its action orderings through plan deordering and reordering, there has been limited focus on the flexibility of executing actions concurrently in a plan. Flexibility of executing actions concurrently, referred to as concurrency, in a POP can be achieved by incorporating action non-concurrency constraints, specifying which actions can not be executed in parallel. This work establishes the necessary and sufficient conditions for non-concurrency constraints between two actions or two subplans with respect to a planning task. We also introduce an algorithm to improve a plan’s concurrency by optimizing resource utilization through substitutions of the plan’s subplans with respect to the corresponding planning task. Our algorithm employs block deordering that eliminates orderings in a POP by encapsulating coherent actions in blocks, and then exploits blocks as candidate subplans for substitutions. Experiments over the benchmark problems from International Planning Competitions (IPC) exhibit considerable improvement in plan concurrency.</p></div>","PeriodicalId":55586,"journal":{"name":"Autonomous Agents and Multi-Agent Systems","volume":"39 1","pages":""},"PeriodicalIF":2.0,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143861407","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Adaptation Procedure in misinformation games 错误信息游戏中的适应程序

IF 2 3区计算机科学

Autonomous Agents and Multi-Agent Systems Pub Date : 2025-03-21 DOI: 10.1007/s10458-025-09704-w

Konstantinos Varsos, Merkouris Papamichail, Giorgos Flouris, Marina Bitsaki

引用次数: 0

On fair and efficient solutions for budget apportionment 关于公平有效的预算分配办法

IF 2 3区计算机科学

Autonomous Agents and Multi-Agent Systems Pub Date : 2025-03-20 DOI: 10.1007/s10458-025-09694-9

Pierre Cardi, Laurent Gourvès, Julien Lesca

{"title":"On fair and efficient solutions for budget apportionment","authors":"Pierre Cardi, Laurent Gourvès, Julien Lesca","doi":"10.1007/s10458-025-09694-9","DOIUrl":"10.1007/s10458-025-09694-9","url":null,"abstract":"<div><p>This article deals with an apportionment problem involving <i>n</i> agents and a common budget <i>B</i>. Each agent submits some demands which are indivisible portions of the budget, and a central authority has to decide which demands to accept. The utility of an agent corresponds to the total amount of her accepted demands. In this context, it is desirable to be fair among the agents and efficient by not wasting the budget. An ideal solution would be to spend exactly <i>B</i>/<i>n</i> for every agent but this is rarely possible because of the indivisibility of the demands. Since combining fairness with efficiency is highly desirable but often impossible, we explore relaxed notions of fairness and efficiency, in order to determine if they go together. Our approach is also constructive because polynomial algorithms that build fair and efficient solutions are also given. The fairness criteria under consideration are the maximization of the minimum agent utility (max–min), proportionality, a customized notion of envy-freeness called jealousy-freeness, and the relaxations up to one or any demand of the previous two concepts. Efficiency in this work is either the maximization of the utilitarian social welfare or Pareto optimality. First we consider fairness and efficiency separately. The existence and computation of solutions that are either fair or efficient are studied. A complete picture of the relations that connect the fairness and efficiency concepts is provided. Second, we determine when fairness and efficiency can be combined for every possible instance. We prove that Pareto optimality is compatible with two notions of fairness, namely max–min and proportionality up to any demand. In contrast, none of the fairness concepts under consideration can be paired with the maximization of utilitarian social welfare. Therefore, we finally conduct a thorough analysis of the price of fairness which bounds the loss of efficiency caused by imposing fairness or one of its relaxations.</p></div>","PeriodicalId":55586,"journal":{"name":"Autonomous Agents and Multi-Agent Systems","volume":"39 1","pages":""},"PeriodicalIF":2.0,"publicationDate":"2025-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143655386","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

“Provably fair” algorithms may perpetuate racial and gender bias: a study of salary dispute resolution “可证明公平”的算法可能使种族和性别偏见永久化：一项关于薪资纠纷解决的研究

IF 2 3区计算机科学

Autonomous Agents and Multi-Agent Systems Pub Date : 2025-03-12 DOI: 10.1007/s10458-025-09703-x

James Hale, Peter H. Kim, Jonathan Gratch

引用次数: 0

Investigating the impact of direct punishment on the emergence of cooperation in multi-agent reinforcement learning systems 研究直接惩罚对多智能体强化学习系统中合作出现的影响

IF 2 3区计算机科学

Autonomous Agents and Multi-Agent Systems Pub Date : 2025-03-11 DOI: 10.1007/s10458-025-09698-5

Nayana Dasgupta, Mirco Musolesi

{"title":"Investigating the impact of direct punishment on the emergence of cooperation in multi-agent reinforcement learning systems","authors":"Nayana Dasgupta, Mirco Musolesi","doi":"10.1007/s10458-025-09698-5","DOIUrl":"10.1007/s10458-025-09698-5","url":null,"abstract":"<div><p>Solving the problem of cooperation is fundamentally important for the creation and maintenance of functional societies. Problems of cooperation are omnipresent within human society, with examples ranging from navigating busy road junctions to negotiating treaties. As the use of AI becomes more pervasive throughout society, the need for socially intelligent agents capable of navigating these complex cooperative dilemmas is becoming increasingly evident. Direct punishment is a ubiquitous social mechanism that has been shown to foster the emergence of cooperation in both humans and non-humans. In the natural world, direct punishment is often strongly coupled with partner selection and reputation and used in conjunction with third-party punishment. The interactions between these mechanisms could potentially enhance the emergence of cooperation within populations. However, no previous work has evaluated the learning dynamics and outcomes emerging from multi-agent reinforcement learning populations that combine these mechanisms. This paper addresses this gap. It presents a comprehensive analysis and evaluation of the behaviors and learning dynamics associated with direct punishment, third-party punishment, partner selection, and reputation. Finally, we discuss the implications of using these mechanisms on the design of cooperative AI systems.</p></div>","PeriodicalId":55586,"journal":{"name":"Autonomous Agents and Multi-Agent Systems","volume":"39 1","pages":""},"PeriodicalIF":2.0,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10458-025-09698-5.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143583433","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0