Weighting Estimation Methods for Opponents' Utility Functions Using Boosting in Multi-Time Negotiations

Abstract

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.

A Negotiating Strategy for a Hybrid Goal Function in Multilateral Negotiation

“…Matsune and Fujita (2018) proposed a method [32] for estimating the utility function of the opponents in a closed multi-issue negotiation. Their method uses the Boosting algorithm, which tries to combine several "weak learners" into one "strong learner" by weighing each one of the learners using the data collected from the agents' offers during the negotiation.…”

Foundations for a Game Theoretic Framework for Agile Acquisition

“…An interesting variant for multi-issue closed negotiations addressing multitime as well as multi-lateral negotiation strategies is described by Matsune and Fujita (2017), who developed not only the concept, but demonstrated it in an agent-based simulation environment. Theoretically, nothing speaks against applying these ideas for acquisition-specific challenges as well, but no applications in this domain within the survey were found.…”