論文

査読有り
2013年

Agent's strategy in multiple-issue negotiation competition and analysis of result

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
  • Shota Morii
  • ,
  • Takayuki Ito

8291
開始ページ
486
終了ページ
493
記述言語
英語
掲載種別
研究論文(国際会議プロシーディングス)
DOI
10.1007/978-3-642-44927-7_38

The Automated Negotiation Agents Competition (ANAC201 2) was organized. It is likely that the strategies of an agent can be applied to real-life negotiation problems. Therefore, this paper deals with a new scenario of multiple-issue negotiation problems. The advantage of the proposed scenario in comparison with existing ones is being close to real-life negotiation problems. In this paper, we present our agent (AgentMRK2) strategy and investigate the effects of strategy that uses the negotiation history in the ANAC2013 qualifying rounds. Moreover, we discuss a change of acquired utility by repeating the negotiations. The study revealed the existence of correlations between the number of bids and the average of utilities. © 2013 Springer-Verlag.

リンク情報
DOI
https://doi.org/10.1007/978-3-642-44927-7_38
ID情報
  • DOI : 10.1007/978-3-642-44927-7_38
  • ISSN : 0302-9743
  • ISSN : 1611-3349
  • SCOPUS ID : 84893128232

エクスポート
BibTeX RIS