Multiagent Negotiation on Multiple Issues with Private Information - Robotics Institute Carnegie Mellon University

Multiagent Negotiation on Multiple Issues with Private Information

Ronghuo Zheng, Nilanjan Chakraborty, Tinglong Dai, and Katia Sycara
Tech. Report, CMU-RI-TR-13-04, Robotics Institute, Carnegie Mellon University, March, 2013

Abstract

In this paper, we present offer generation methods for negotiation among multiple agents on multiple issues where agents have no knowledge about the preferences of other agents. Most of the existing literature on non-mediated negotiation consider agents with either full information or probabilistic beliefs about the other agents preferences on the issues. However, in reality, it is usually not possible for agents to have complete information about other agents preferences or accurate probability distributions. Moreover, the extant literature typically assumes linear utility functions. We present a reactive offer generation method for general multiagent multi-attribute negotiation, where the agents have non-linear utility functions and no information about the utility functions of other agents. We prove the convergence of the proposing method to an agreement acceptable to the agents. We also prove that rational agents do not have any incentive to deviate from the proposed strategy. We further present simulation results to demonstrate that on randomly generated problem instances the negotiation solution obtained by using our strategy is quite close to the Nash bargaining solution.

BibTeX

@techreport{Zheng-2013-7676,
author = {Ronghuo Zheng and Nilanjan Chakraborty and Tinglong Dai and Katia Sycara},
title = {Multiagent Negotiation on Multiple Issues with Private Information},
year = {2013},
month = {March},
institute = {Carnegie Mellon University},
address = {Pittsburgh, PA},
number = {CMU-RI-TR-13-04},
keywords = {multiagent coordination, negotiation, distributed decision making.},
}