Recent Advances in Agent-based Negotiation: Formal Models and Human Aspects

This document was uploaded by one of our users. The uploader already confirmed that they had the permission to publish it. If you are author/publisher or own the copyright of this documents, please report to us by using this DMCA report form.

Simply click on the Download Book button.

Yes, Book downloads on Ebookily are 100% Free.

Sometimes the book is free on Amazon As well, so go ahead and hit "Search on Amazon"

This volume comprises carefully selected and reviewed outcomes of the 12th International Workshop on Automated Negotiations (ACAN) held in Macao, 2019, in conjunction with International Joint Conference on Artificial Intelligence (IJCAI) 2019. It focuses on human aspects of automated negotiation and the recent advances in negotiation frameworks and strategies. Written by leading academic and industrial researchers, it is a valuable resource for professionals and scholars working on complex automated negotiations.

Author(s): Reyhan Aydoğan, Takayuki Ito, Ahmed Moustafa, Takanobu Otsuka, Minjie Zhang
Series: Studies in Computational Intelligence, 958
Publisher: Springer
Year: 2021

Language: English
Pages: 131
City: Singapore

Preface
Contents
Editors and Contributors
Human Aspects in Agent-based Negotiation
Effect of Awareness of Other Side's Gain on Negotiation Outcome, Emotion, Argument, and Bidding Behavior
1 Introduction
2 Related Work
3 Human–Human Negotiation
3.1 Awareness of Opponent's Gain
3.2 Emotions in Negotiation
3.3 Argumentation in Negotiation
4 Bidding Behavior
5 Structured Human Negotiations
6 Experimental Evaluation
6.1 Experimental Setup
6.2 Analysis of Negotiation Outcome
6.3 Analysis of Arguments
6.4 Analysis of Bidding Behavior
6.5 Analysis of Emotion
6.6 Analysis of Questionnaire
7 Conclusion and Future Work
References
Facilitation in Abstract Argumentation with Abstract Interpretation
1 Introduction
1.1 Abstract Interpretation as Facilitation
2 Technical Preliminaries
2.1 Abstract Argumentation
2.2 Order and Galois Connection for Abstract Interpretation
3 Argumentation Frameworks for Abstraction
3.1 Lattices
3.2 Abstraction and Concretisation
3.3 Computation of Abstract Space Argumentation Frameworks from a Concrete Space Argumentation Framework
3.4 Preferred Sets in Concrete and Abstract Spaces
3.5 Comparisons to Dung Preferred Semantics and cf2 Semantics, and Observations
4 Conclusion with Related Work
4.1 Related Work
4.2 Conclusion
References
How to Recognize and Explain Bidding Strategies in Negotiation Support Systems
1 Introduction
2 Related Work
3 Typical Bidding Strategies
4 Bids, Utilities and Moves
4.1 Optimal Bidding Strategy Recognition
5 Expectations and Aberrations
6 Generating Explanations
7 Evaluation
7.1 Preparation
7.2 Conditions
7.3 Metrics
8 Results
8.1 Pilot 1
8.2 Pilot 2
8.3 Full Experiment
9 Conclusion
References
Negotiation Frameworks, Strategies, and Recommenders
NegMAS: A Platform for Situated Negotiations
1 Introduction
2 Situated Negotiations
3 System Design
3.1 Issues and Outcomes
3.2 Utility Functions
3.3 Negotiators
3.4 Controllers
3.5 Agents
3.6 Mechanisms
3.7 Worlds
4 Tools and Common Components
5 Applications: Focus on SCML
6 Using NegMAS for Developing SCM Agents
7 Conclusions
References
Can a Reinforcement Learning Trading Agent Beat Zero Intelligence Plus at Its Own Game?
1 Introduction
2 Trading Agents
2.1 ZI Agents
2.2 ZIP Agents
2.3 RL Agents (ZIQ+)
3 Setup
4 Results
5 Conclusions
References
Negotiation in Hidden Identity: Designing Protocol for Werewolf Game
1 Introduction
2 Background
2.1 Werewolf Game: Hidden Identity in Communication Game
2.2 Game AI Studies: From Chess to Werewolf
3 Model of Werewolf Games
3.1 Basic Rule on Close-Rule Werewolf Game
3.2 Lack of Objective Resources
3.3 Reasoning for Modeling the Intentions of Others
3.4 Persuasion as Modeling Self from the Perspective of Others
3.5 Requirements for a Werewolf Game Protocol
4 Werewolf Game Protocol
4.1 Word
4.2 Sentence
4.3 Operator
4.4 Grammar Notes
4.5 About Omitting Subjects (UNSPEC)
4.6 Example Sentences
5 Conclusion
References
Multi-Agent Recommender System
1 Introduction
2 Literature Review
2.1 Machine Learning (ML)
2.2 Recommender Systems (RS)
3 The MARS Recommender System
3.1 MARS Architecture
3.2 Training and Test Data
3.3 The Manager Agent
4 Experimental Evaluation
4.1 Dataset
4.2 Data Pre-processing
4.3 Implementation Environment
4.4 Evaluation Metrics
5 Performance Analysis of MARS
5.1 Performance of the Components MARS
5.2 A Comparison of MARS with Other Systems
6 Conclusion and Future Work
References