Belief Revision in Multi-Agent Systems

The ability to respond sensibly to changing and conflicting beliefs is an integral part of intelligent agency. To this end, we outline the design and implementation of a Distributed Assumption-based Truth Maintenance System (DATMS) appropriate for controlling cooperative problem solving in a dynamic real world multi-agent community. Our DATMS works on the principle of local coherence which means that different agents can have different perspectives on the same fact provided that these stances are appropriately justified. The belief revision algorithm is presented, the meta-level code needed to ensure that all system-wide queries can be uniquely answered is described, and the DATMS’ implementation in a general purpose multi-agent shell is discussed.

Convergence and knowledge processing in multi-agent systems

Multi-agent systems are complex systems comprised of multiple intelligent agents that act either independently or in cooperation with one another. Agent-based modelling is a method for studying complex systems like economies, societies, ecologies etc. Due to their complexity, very often mathematical analysis is limited in its ability to analyse such systems. In this case, agent-based modelling offers a practical, constructive method of analysis. The objective of this book is to shed light on some emergent properties of multi-agent systems. The authors focus their investigation on the effect of knowledge exchange on the convergence of complex, multi-agent systems.

Formalization of Interaction Events in Multi-agent Systems

The problem of the description of interaction between spatially divided agents in the form of dialogues is explored. The concept of processes synchronization is analyzed to formalize the specification of interaction at the level of events constituting the processes. The approach to formalization of the description of conditions of synchronization when both the independent behavior and the communications of agents can be presented at a logic level is offered. It is shown, that the collective behavior of agents can be specified by the synthetic temporal logic that unites linear and branching time temporal logics.

Organisational adaptation of multi-agent systems in a peer-to-peer scenario

Organisations in Multi-Agent Systems (MAS) have proven to be successful in regulating agent societies. Nevertheless, changes in agents' behaviour or in the dynamics of the environment may lead to a poor fulfilment of the system's purposes, and so the entire organisation needs to be adapted. In this paper we focus on endowing the organisation with adaptation capabilities, instead of expecting agents to be capable of adapting the organisation by themselves. We regard this organisational adaptation as an assisting service provided by what we call the Assistance Layer. Our generic Two Level Assisted MAS Architecture (2-LAMA) incorporates such a layer. We empirically evaluate this approach by means of an agent-based simulator we have developed for the P2P sharing network domain. This simulator implements 2-LAMA architecture and supports the comparison between different adaptation methods, as well as, with the standard BitTorrent protocol. In particular, we present two alternatives to perform norm adaptation and one method to adapt agents'relationships. The results show improved performance and demonstrate that the cost of introducing an additional layer in charge of the system's adaptation is lower than its benefits.

Experience generalization for multi-agent reinforcement learning

On-line learning methods have been applied successfully in multi-agent systems to achieve coordination among agents. Learning in multi-agent systems implies in a non-stationary scenario perceived by the agents, since the behavior of other agents may change as they simultaneously learn how to improve their actions. Non-stationary scenarios can be modeled as Markov Games, which can be solved using the Minimax-Q algorithm a combination of Q-learning (a Reinforcement Learning (RL) algorithm which directly learns an optimal control policy) and the Minimax algorithm. However, finding optimal control policies using any RL algorithm (Q-learning and Minimax-Q included) can be very time consuming. Trying to improve the learning time of Q-learning, we considered the QS-algorithm. in which a single experience can update more than a single action value by using a spreading function. In this paper, we contribute a Minimax-QS algorithm which combines the Minimax-Q algorithm and the QS-algorithm. We conduct a series of empirical evaluation of the algorithm in a simplified simulator of the soccer domain. We show that even using a very simple domain-dependent spreading function, the performance of the learning algorithm can be improved.

Multi-agent systems and birtual producers in electronic marketplaces

This paper presents an agent-based simulator designed for analyzing agent market strategies based on a complete understanding of buyer and seller behaviours, preference models and pricing algorithms, considering user risk preferences. The system includes agents that are capable of improving their performance with their own experience, by adapting to the market conditions. In the simulated market agents interact in several different ways and may joint together to form coalitions. In this paper we address multi-agent coalitions to analyse Distributed Generation in Electricity Markets

Case-based reasoning for meta-heuristics self-parameterization in a multi-agent scheduling system

A novel agent-based approach to Meta-Heuristics self-configuration is proposed in this work. Meta-heuristics are examples of algorithms where parameters need to be set up as efficient as possible in order to unsure its performance. This paper presents a learning module for self-parameterization of Meta-heuristics (MHs) in a Multi-Agent System (MAS) for resolution of scheduling problems. The learning is based on Case-based Reasoning (CBR) and two different integration approaches are proposed. A computational study is made for comparing the two CBR integration perspectives. In the end, some conclusions are reached and future work outlined.

Argumentation as Distributed Belief Revision: Conflict Resolution in Decentralised Multi-agent Systems

Decentralised co-operative multi-agent systems are computational systems where conflicts are frequent due to the nature of the represented knowledge. Negotiation methodologies, in this case argumentation based negotiation methodologies, were developed and applied to solve unforeseeable and, therefore, unavoidable conflicts. The supporting computational model is a distributed belief revision system where argumentation plays the decisive role of revision. The distributed belief revision system detects, isolates and solves, whenever possible, the identified conflicts. The detection and isolation of the conflicts is automatically performed by the distributed consistency mechanism and the resolution of the conflict, or belief revision, is achieved via argumentation. We propose and describe two argumentation protocols intended to solve different types of identified information conflicts: context dependent and context independent conflicts. While the protocol for context dependent conflicts generates new consensual alternatives, the latter chooses to adopt the soundest, strongest argument presented. The paper shows the suitability of using argumentation as a distributed decentralised belief revision protocol to solve unavoidable conflicts.