Learning Other Agents' Preferences in Multi-Agent Negotiation Using the Bayesian Classifier

Bui, Hung; Venkatesh, Svetha; Kieronska, D.

doi:10.1142/s0218843099000149

Cited by 43 publications

(51 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When agents are capable of learning about other agents' task-solving abilities, communication is reduced from broadcasting to everyone to communicating exact messages to only those agents that have high probabilities to win the bids for those tasks. A related approach is presented in [37,36]: here, Bayesian learning is used to incrementally update models of other agents to reduce communication load by anticipating their future actions based on their previous ones. Case-based learning has also been used to develop successful joint plans based on one's historical expectations of other agents' actions [83].…”

Section: Teammate Modelingmentioning

confidence: 99%

Cooperative Multi-Agent Learning: The State of the Art

Panait

Luke

2005

Auton Agent Multi-Agent Syst

987

559

View full text Add to dashboard Cite

Cooperative multi-agent systems are ones in which several agents attempt, through their interaction, to jointly solve tasks or to maximize utility. Due to the interactions among the agents, multi-agent problem complexity can rise rapidly with the number of agents or their behavioral sophistication. The challenge this presents to the task of programming solutions to multi-agent systems problems has spawned increasing interest in machine learning techniques to automate the search and optimization process.We provide a broad survey of the cooperative multi-agent learning literature. Previous surveys of this area have largely focused on issues common to specific subareas (for example, reinforcement learning or robotics). In this survey we attempt to draw from multi-agent learning work in a spectrum of areas, including reinforcement learning, evolutionary computation, game theory, complex systems, agent modeling, and robotics.We find that this broad view leads to a division of the work into two categories, each with its own special issues: applying a single learner to discover joint solutions to multi-agent problems (team learning), or using multiple simultaneous learners, often one per agent (concurrent learning). Additionally, we discuss direct and indirect communication in connection with learning, plus open issues in task decomposition, scalability, and adaptive dynamics. We conclude with a presentation of multi-agent learning problem domains, and a list of multi-agent learning resources.

show abstract

Section: Teammate Modelingmentioning

confidence: 99%

Cooperative Multi-Agent Learning: The State of the Art

Panait

Luke

2005

Auton Agent Multi-Agent Syst

987

559

View full text Add to dashboard Cite

show abstract

“…Yet, this area abounds in methods for learning models of particular individuals' strategies (cf. [2,5,3,6,15]). Therefore, proposing new OM methods was not an issue by itself in the development of ADHOC.…”

Section: Adhocmentioning

confidence: 99%

“…ADHOC assumes that interaction takes place between only two agents at a time in discrete encounters 2 Any encounter can be interpreted as a fixed length iterated two-player normal-form game [7]; however, the OM method we use in our implementation does not require that…”

Section: Overviewmentioning

confidence: 99%

See 1 more Smart Citation

Multiagent Learning for Open Systems: A Study in Opponent Classification

Rovatsos

Weiß

Wolf

2003

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Open systems are becoming increasingly important in a variety of distributed, networked computer applications. Their characteristics, such as agent diversity, heterogeneity and fluctuation, confront multiagent learning with new challenges. This paper presents the interaction learning meta-architecture InFFrA as one possible answer to these challenges, and introduces the opponent classification heuristic ADHOC as a concrete multiagent learning method that has been designed on the basis of InFFrA. Extensive experimental validation proves the adequacy of ADHOC in a scenario of iterated multiagent games and underlines the usefulness of schemas such as InFFrA specifically tailored for open multiagent learning environments. At the same time, limitations in the performance of ADHOC suggest further improvements to the methods used here. Also, the results obtained from this study allow more general conclusions regarding the problems of learning in open systems to be drawn.

show abstract

“…Despite the fact that automated negotiation has been studied since the 90's decade (18,19,20,21), there is still a wide range of problems whose solution has not been treated in the literature.…”

Section: Motivationmentioning

confidence: 99%

Complex negotiations in multi-agent systems.

Anguix¹

View full text Add to dashboard Cite

Multi-agent systems (MAS) are distributed systems where autonomous entities called agents, either human or software, pursue their own goals. The MAS paradigm has been proposed as the appropriate modeling approach for the deployment of applications like electronic commerce, multi-robot systems, security applications, and so forth. In the MAS community, the vision of open multi-agent system, where heterogeneous agents can enter and leave the system dynamically, has gained strength as a potentially interesting modeling paradigm due to its conceptual relation with technologies like world wide web, grid computing, and virtual organizations. Given the heterogeneity and agent's self-interest, conflict is a candidate phenomenon to arise in multi-agent systems.In the last few years, the term agreement technologies has been used to address all the mechanisms that, directly or indirectly, promote the resolution of conflicts in computational systems like multi-agent systems. Among agreement technologies, automated negotiation is proposed as one key mechanism in conflict resolution due to its analogous use in human conflict resolution. Automated negotiation consists of an automated exchange of proposals carried out by software agents on behalf of their users. The final goal is the achievement of an agreement with all the involved parts.Despite being studied by scholars in Artificial Intelligence for several years, several problems have not been addressed by the scientific community yet. The main objective of this thesis is proposing negotiation models for complex scenarios where the complexity may stem from (i) limited computational capabilities or (ii) the necessity to accommodate the preferences of multiple individuales. In the first part of the thesis we propose a bilateral negotiation model for the problem of negotiation in Ambient Intelligence (AmI), a domain with a special emphasis iii on computational efficiency due to the limited capability of AmI devices. In the second part of the thesis we propose several negotiation models for agent-based negotiation teams. A negotiation team is a group of individuals that acts together as single negotiation party due to its common interests in the negotiation at hand.The complexity of negotiation teams resides in the fact that despite having common interests, intra-team conflict is also present. As far as we are concerned, the topic of negotiation teams in MAS is introduced with this thesis.iv ResumenLos sistemas multi-agente (SMA) son sistemas distribuidos donde entidades autó-nomas llamadas agentes, ya sean humanos o software, persiguen sus propios objetivos. El paradigma de SMA ha sido propuesto como la aproximación de modelo apropiada para aplicaciones como el comercio electrónico, los sistemas multi-robot, aplicaciones de seguridad, etc. En la comunidad de SMA, la visión de sistemas multi-agente abiertos, donde agentes heterogéneos pueden entrar y salir del sistema dinámicamente, ha cobrado fuerza como paradigma de modelado debido a su relación conceptual con tecnolo...

show abstract

Learning Other Agents' Preferences in Multi-Agent Negotiation Using the Bayesian Classifier

Cited by 43 publications

References 6 publications

Cooperative Multi-Agent Learning: The State of the Art

Cooperative Multi-Agent Learning: The State of the Art

Multiagent Learning for Open Systems: A Study in Opponent Classification

Complex negotiations in multi-agent systems.

Contact Info

Product

Resources

About