Meta-reasoning for Agents’ Private Knowledge Detection

sis.uncc.edu

Abstract. In this paper, we consider the role of meta-reasoning in achieving effective coordination among multiple agents in maintaining and executing joint plans in an uncertain environment. We assume that each agent has responsibility for performing a particular set ...

Drawing on on our work in the area of Distributed Arti cial Intelligence, we propose the rudiments of a view of multiagent reasoning that relates current philosophical intuitions, theoretical foundations, and preliminary implementation. The philosophical position we take is a combination of Daniel Dennett's philosophy of the ladder of personhood (consisting of rationality, intentionality, stance, reciprocity, communication, and consciousness) on one hand, and the utilitarian philosophy of sel sh utility maximization on the other hand. The theories we incorporate are logics of knowledge and belief, which in addressing the multiagent issues can be developed based on a recursive version of the Kripke structure, and the related elds of utility, decision and game theories. Our preliminary implementation, the Recursive Modeling Method (RMM), lets an agent coordinate its actions with the actions of other agents, cooperate with them when appropriate, and rationally choose an optimal form of communication with them.

2008

We review the state of the art in knowledge representation formalisms for multi-agent systems. We divide work in this area into two categories. In the first category are approaches that attempt to represent the cognitive state of rational agents, and to characterize logically how such a state leads a rational agent to act. We begin by motivating this approach. We then describe four of the best-known such logical frameworks, and discuss the possible roles that such logics can play in helping us to engineer artificial agents. In the second category are approaches based on representing the strategic structure of a multi-agent environment, and in particular, the powers that agents have, either individually or in coalitions. Here, we describe Coalition Logic, Alternatingtime Temporal Logic (ATL), and epistemic extensions.

International Journal of Information Systems for Crisis Response and Management, 2017

The disaster response still faces problems of collaboration due to lack of policies concerning the information exchange during the response. Moreover, plans are prepared to respond to a disaster, but drills to apply them are limited and do not allow to determine their efficiency and conflicts with other organizations. This paper presents a framework allowing for different organizations involving in the disaster response to assess their collaboration through its simulation using an explicit representation of their knowledge. This framework is based on a multi-agent system composed of three generic agent models to represent the organizational structure of disaster response. The decision-making about response actions is done through task decomposition and repartition. It is based reasoning on ontologies which provides an explicit trace of the response plans design and their execution. Such framework aims at identifying cooperation problems and testing strategies of information exchange...

iSys - Brazilian Journal of Information Systems, 2017

In this work we compile our research regarding meta-information in multi-agent systems. In particular, we describe some agents profiles represent- ing different attitudes which describe how agents consider meta-information in their decisions-making and reasoning processes. Furthermore, we describe how we have combined different meta-information available in multi-agent systems with an argumentation-based reasoning mechanism. In our approach, agents are able to decide more conflicts between information/arguments, given that they are able to use different meta-information (often combined) to decide between such conflicting information. Our framework for meta-information in multi- agent systems was implemented based on a modular architecture, thus other meta-information can be added, as well as different meta-information can be combined in order to create new agents profiles. Therefore, in our approach, different agents profiles can be instantiated for different application domains, al...

Proc. ISCRAM, 2006

Natural and human-made disasters create unparalleled challenges to Disaster Situation Management (DSM). One of the major weaknesses of the current DSM solutions is the lack of comprehensive understanding of the overall disaster operational situation, and very often making decisions based on a single event. Such weakness is clearly exhibited by the solutions based on the widely used Belief-Desire-Intention (BDI) models for building the Muiti-Agent Systems (MAS). In this work we present the adaptation of the AESOP situation management architecture to address the requirements of disaster relief operations. In particular, we extend the existing BDI model with the capability of situation awareness. We describe how the key functions of event collection, situation identification, and situation assessment are implemented in MAS architecture suitable to the characteristics of large-scale disaster recovery. We present the details of a BDI agent in this architecture including a skeleton ontology, and the distributed service architecture of the AESOP platform.

2004

Deliberative agents operating in open environments must make complex real-time decisions on scheduling and coordination of domain activities. These decisions are made in the context of limited resources and uncertainty about the outcomes of activities. We describe a reinforcement learning based approach for efficient meta-level reasoning. Empirical results showing the effectiveness of meta-level reasoning in a complex domain are provided.

2007

Traditional work in belief revision deals with the question of what an agent should believe upon receiving new information. We will give an overview about what can be concluded about an agent based on an observation of its belief revision behaviour. The observation contains partial information about the revision inputs received by the agent and its beliefs upon receiving them. We will sketch a method for reasoning about past and future beliefs of the agent and predicting which inputs it accepts and rejects. The focus of this talk will be on different degrees of incompleteness of the observation and variants of the general question we are able to deal with.

Knowledge Based Systems, 2009

This paper presents analysis and simulation of meta-reasoning processes based on an agent-based meta-level architecture for strategic reasoning in naval planning. The architecture was designed as a generic agent model and instantiated with decision knowledge acquired from naval domain experts and was specified as an executable agent-based model which has been used to perform a number of simulations. To evaluate the simulation results, relevant properties for the planning decision were identified and formalized. These properties have been validated for the simulation traces.

2008

ALADDIN [1] is a multidisciplinary project that is developing novel techniques, architectures, and mechanisms for multi-agent systems in uncertain and dynamic environments. The application focus of the project is disaster management. Research within a number of themes is being pursued and this is considering different aspects of the interaction between autonomous agents and the decentralised system architectures that support those interactions. The aim of the research is to contribute to building more robust multi-agent systems for future applications in disaster management and other similar domains. This paper summarises the four research themes that make up the ALADDIN project-Applications, Decentralised System Architectures, Individual Agents, and Multiple Agents.