Weighting estimation methods for opponents' utility functions using boosting in multi-time negotiations
Weighting estimation methods for opponents' utility functions using boosting in multi-time negotiations
复制标题
多次谈判中使用boosting的对手效用函数加权估计方法
DOI：
10.1109/agents.2017.8015296
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
Fujita Katsuhide
中科院分区：
文献类型：
--
作者：
Matsune Takaki;Fujita Katsuhide
近年来，多问题闭合谈判在多智能体系统中引起了人们的关注。特别是，多时间和多边谈判战略是多议题闭门谈判的重要议题。在多问题闭合谈判中，由于对手的效用信息是不公开的，自动协商主体需要通过学习对手的行为来估计对手的效用函数。然而，由于以下原因，估计对手的效用函数是困难的：(1)无法获得估计对手效用函数的训练数据集。(2)学习模型难以应用于不同的谈判域和对手。本文提出了一种基于最小二乘法和非线性规划的Boosting估计对手效用函数的新方法。该方法对已有的几种效用函数估计方法所估计的效用函数进行加权，并将每个加权函数相加输出改进的效用函数。现有的使用Boosting的方法是基于频率的方法，该方法计算提供的值的数量，考虑到提供值所经过的时间。实验结果表明，与现有的无Booost的效用函数估计方法相比，在不同的条件下，估计对手效用函数的精度都有明显的提高。
Recently, multi-issue closed negotiations have attracted attention in multi-agent systems. In particular, multi-time and multilateral negotiation strategies are important topics in multi-issue closed negotiations. In multi-issue closed negotiations, an automated negotiating agent needs to have strategies for estimating an opponent's utility function by learning the opponent's behaviors since the opponent's utility information is not open to others. However, it is difficult to estimate an opponent's utility function for the following reasons: (1) Training datasets for estimating opponents' utility functions cannot be obtained. (2) It is difficult to apply the learned model to different negotiation domains and opponents. In this paper, we propose a novel method of estimating the opponents' utility functions using boosting based on the least-squares method and nonlinear programming. Our proposed method weights each utility function estimated by several existing utility function estimation methods and outputs improved utility function by summing each weighted function. The existing methods using boosting are based on the frequency-based method, which counts the number of values offered, considering the time elapsed when they offered. Our experimental results demonstrate that the accuracy of estimating opponents' utility functions is significantly improved under various conditions compared with the existing utility function estimation methods without boosting.
登录
查看更多内容
DOI：
--
发表时间：
2014
期刊：
影响因子：
--
作者：
Masanori Ikarashi and Katsuhide Fujita
通讯作者：
Masanori Ikarashi and Katsuhide Fujita
DOI：
--
发表时间：
2014
期刊：
影响因子：
--
作者：
Kenji Kakimoto;Masao Yamagishi;Isao Yamada;Katsuhide Fujita
通讯作者：
Katsuhide Fujita
基于代理的复杂自动协商的现代方法
DOI：
--
发表时间：
2017
期刊：
影响因子：
--
作者：
Katsuhide Fujita;Quan Bai;Takayuki Ito;Minjie Zhang;Fenghui Ren;Reyhan Aydogan;Rafik Hadfi
通讯作者：
Rafik Hadfi
DOI：
10.1007/978-3-642-24696-8_7
发表时间：
2012
期刊：
--
影响因子：
--
作者：
T. Baarslag;K. Hindriks;C. Jonker;Sarit Kraus;R. Lin
通讯作者：
T. Baarslag;K. Hindriks;C. Jonker;Sarit Kraus;R. Lin
DOI：
10.1007/978-3-319-51563-2_10
发表时间：
2017
期刊：
Studies in computational intelligence
影响因子：
--
作者：
Reyhan Aydoğan;David Festen;K. Hindriks;C. Jonker
通讯作者：
C. Jonker