多智能體系統(tǒng)可以看作由多個具有自主決策能力的軟件智能體組成，各智能體之間會直接或間接地相互作用和影響。通�？梢园讯嘀悄荏w系統(tǒng)分為兩大類：合作式多智能體系統(tǒng)和非合作式多智能體系統(tǒng)，前者研究的核心問題是各智能體如何利用有限的局部信息，通過自主學(xué)習(xí)有效協(xié)作達到最優(yōu)的共同目標；而后者一個重要問題是如何采用有效激勵機制，促使各智能體主動協(xié)調(diào)合作，從而最大化系統(tǒng)整體性能。
1 Introduction
1.1 Overview of the Chapters
1.2 Guide to the Book
References
2 Background and Previous Work
2.1 Background
2.1.1 Single-Shot Normal-Form Game
2.1.2 Repeated Games
2.2 Cooperative Multiagent Systems
2.2.1 Achieving Nash Equilibrium
2.2.2 Achieving Fairness
2.2.3 Achieving Social Optimality
2.3 Competitive Multiagent Systems
2.3.1 Achieving Nash Equilibrium
2.3.2 Maximizing Individual Benefits
2.3.3 Achieving Pareto-Optimality
References
3 Fairness in Cooperative Multiagent Systems
3.1 An Adaptive Periodic Strategy for Achieving Fairness
3.1.1 Motivation
3.1.2 Problem Specification
3.1.3 An Adaptive Periodic Strategy
3.1.4 Properties of the Adaptive Strategy
3.1.5 Experimental Evaluations
3.2 Game-Theoretic Fairness Models
3.2.1 Incorporating Fairness into Agent Interactions
Modeled as Two-Player Normal-Form Games
3.2.2 Incorporating Fairness into Infinitely Repeated
Games with Conflicting Interests for Conflict Elimination
References
4 Social Optimality in Cooperative Multiagent Systems
4.1 Reinforcement Social Learning of Coordination
in Cooperative Games
4.1.1 Social Learning Framework
4.1.2 Experimental Evaluations
4.2 Reinforcement Social Learning of Coordination
in General-Sum Games
4.2.1 Social Learning Framework
4.2.2 Analysis of the Learning Performance Under
the Social Learning Framework
4.2.3 Experimental Evaluations
4.3 Achieving Socially Optimal Allocations Through Negotiation
4.3.1 Multiagent Resource Allocation Problem
Through Negotiation
4.3.2 The APSOPA Protocol to Reach Socially Optimal
Allocation
4.3.3 Convergence of APSOPA to Socially Optimal Allocation..
4.3.4 Experimental Evaluation
References
5 Individual Rationality in Competitive Multiagent Systems
5.1 Introduction
5.2 Negotiation Model
5.3 ABiNeS: An Adaptive Bilateral Negotiating Strategy
5.3.1 Acceptance-Threshold (AT) Component
5.3.2 Next-Bid (NB) Component
5.3.3 Acceptance-Condition (AC) Component
5.3.4 Termination-Condition (TC) Component
5.4 Experimental Simulations and Evaluations
5.4.1 Experimental Settings
5.4.2 Experimental Results and Analysis: Efficiency
5.4.3 Detailed Analysis of ABiNeS Strategy
5.4.4 The Empirical Game-Theoretic Analysis: Robustness
5.5 Conclusion
References
6 Social Optimality in Competitive Multiagent Systems
6.1 Achieving Socially Optimal Solutions in the Context
of Infinitely Repeated Games
6.1.1 Learning Environment and Goal
6.1.2 TaFSO: A Learning Approach Toward SOSNE Outcomes:
6.1.3 Experimental Simulations
6.2 Achieving Socially Optimal Solutions in the Social
Learning Framework
6.2.1 Social Learning Environment and Goal
6.2.2 Learning Framework
6.2.3 Experimental Simulations
References
7 Conclusion
Reference
A The 57 Structurally Distinct Games
書單推薦
