상세정보

Distributed Constraint Satisfaction gives an overview of Constraint Satisfaction Problems (CSPs), adapts related search algorithms and consistency algorithms for applications to multi-agent systems, and consolidates recent research devoted to cooperation in such systems. The techniques introduced are applied to various problems in multi-agent systems. Among the new approaches is a hybrid-type algorithm for weak-commitment search combining backtracking and iterative improvement. Also, an extension of the basic CSP formalization called "Partial CSP" is introduced in order to handle over-constrained CSPs.

When multiple agents are in a shared environment, there usually exist con­ straints among the possible actions of these agents. A distributed constraint satisfaction problem (distributed CSP) is a problem in which the goal is to find a consistent combination of actions that satisfies these inter-agent constraints. More specifically, a distributed CSP is a constraint satisfaction problem (CSP) in which multiple agents are involved. A constraint satisfaction problem is a problem in which the goal is to find a consistent assignment of values to variables. Even though the definition of a CSP is very simple, a surprisingly wide variety of artificial intelligence (AI) problems can be formalized as CSPs. Therefore, the research on CSPs has a long and distinguished history in AI (Mackworth 1992; Dechter 1992; Tsang 1993; Kumar 1992). A distributed CSP is a CSP in which variables and constraints are distributed among multiple autonomous agents. Various application problems in Multi-agent Systems (MAS) that are concerned with finding a consistent combination of agent actions can he formalized as dis­ tributed CSPs. Therefore, we can consid(~r distributed CSPs as a general framework for MAS, and algorithms for solving distributed CSPs as impor­ tant infrastructures for cooperation in MAS. This book gives an overview of the research on distributed CSPs, as well as introductory material on CSPs. In Chapter 1. we show the problem defi­ nition of normal, centralized CSPs and describe algorithms for solving CSPs.

정보제공 :

1. Constraint Satisfaction Problem.- 1.1 Introduction.- 1.2 Problem Definition.- 1.3 Algorithms for Solving CSPs.- 1.3.1 Backtracking.- 1.3.2 Iterative Improvement.- 1.3.3 Consistency Algorithms.- 1.4 Hybrid-Type Algorithm of Backtracking and Iterative Improvement.- 1.4.1 Weak-Commitment Search Algorithm.- 1.4.2 Example of Algorithm Execution.- 1.4.3 Evaluations.- 1.4.4 Algorithm Complexity.- 1.5 Analyzing Landscape of CSPs.- 1.5.1 Introduction.- 1.5.2 Hill-Climbing Algorithm.- 1.5.3 Analyzing State-Space.- 1.5.4 Discussions.- 1.6 Partial Constraint Satisfaction Problem.- 1.6.1 Introduction.- 1.6.2 Formalization.- 1.6.3 Algorithms.- 1.7 Summary.- 2. Distributed Constraint Satisfaction Problem.- 2.1 Introduction.- 2.2 Problem Formalization.- 2.3 Application Problems.- 2.3.1 Recognition Problem.- 2.3.2 Allocation Problem.- 2.3.3 Multi-agent Truth Maintenance.- 2.3.4 Time-Tabling/Scheduling Tasks.- 2.4 Classification of Algorithms for Solving Distributed CSPs.- 2.5 Summary.- 3. Asynchronous Backtracking.- 3.1 Introduction.- 3.2 Assumptions.- 3.3 Simple Algorithms.- 3.3.1 Centralized Method.- 3.3.2 Synchronous Backtracking.- 3.4 Asynchronous Backtracking Algorithm.- 3.4.1 Overview.- 3.4.2 Characteristics of the Asynchronous Backtracking Algorithm.- 3.4.3 Example of Algorithm Execution.- 3.4.4 Algorithm Soundness and Completeness.- 3.5 Evaluations.- 3.6 Summary.- 4. Asynchronous Weak-Commitment Search.- 4.1 Introduction.- 4.2 Basic Ideas.- 4.3 Details of Algorithm.- 4.4 Example of Algorithm Execution.- 4.5 Algorithm Completeness.- 4.6 Evaluations.- 4.7 Summary.- 5. Distributed Breakout.- 5.1 Introduction.- 5.2 Breakout Algorithm.- 5.3 Basic Ideas.- 5.4 Details of Algorithm.- 5.5 Example of Algorithm Execution.- 5.6 Evaluations.- 5.7 Discussions.- 5.8 Summary.- 6. Distributed Consistency Algorithm.- 6.1 Introduction.- 6.2 Overview of Distributed ATMS.- 6.2.1 ATMS.- 6.2.2 Distributed ATMS.- 6.3 Distributed Consistency Algorithm Using Distributed ATMS..- 6.4 Example of Algorithm Execution.- 6.5 Evaluations.- 6.6 Summary.- 7. Handling Multiple Local Variables.- 7.1 Introduction.- 7.2 Agent-Prioritization Approach.- 7.3 Asynchronous Weak-Commitment Search with Multiple Local Variables.- 7.3.1 Basic Ideas.- 7.3.2 Details of Algorithm.- 7.3.3 Example of Algorithm Execution.- 7.4 Evaluations.- 7.5 Summary.- 8. Handling Over-Constrained Situations.- 8.1 Introduction.- 8.2 Problem Formalization.- 8.3 Distributed Maximal CSPs.- 8.3.1 Problem Formalization.- 8.3.2 Algorithms.- 8.3.3 Evaluations.- 8.4 Distributed Hierarchical CSPs.- 8.4.1 Problem Formalization.- 8.4.2 Asynchronous Incremental Relaxation.- 8.4.3 Example of Algorithm Execution.- 8.4.4 Algorithm Completeness.- 8.4.5 Evaluations.- 8.5 Summary.- 9. Summary and Future Issues.

정보제공 :