Optimal Negotiation Strategies for Agents with Incomplete Information

Abstract

This paper analyzes the process of automated negotiation between two competitive agents that have firm deadlines and incomplete information about their opponent. Generally speaking, the outcome of a negotiation depends on many parameters-including the agents' preferences, their reservation limits, their attitude toward time and the strategies they use. Although in most realistic situations it is not possible for agents to have complete information about each of these parameters for its opponent, it is not uncommon for agents to have partial information about some of them. Under such uncertainty, our aim is to determine how an agent can exploit its available information to select an optimal strategy. Here, in particular, the optimal strategies are determined considering all possible ways in which time can effect negotiation. Moreover, we list the conditions for convergence when both agents use their respective optimal strategies and study the effect of time on negotiation outcome.

Towards a Quantitative Concession-Based Classification Method of Negotiation Strategies

“…The alternating-offers protocol is supplemented with a deadline D at the end of the time line, so for any t ∈ T , we stipulate that the deadline has been reached when t = D, at which moment both agents receive utility 0. This is the same setup as [6], with the exception that issues are not necessarily realvalued and both agents have the same deadline. We represent by x t A→B the negotiation outcome proposed by agent A to agent B at time t. A negotiation thread (cf.…”

“…Given a negotiation thread H t A↔B between agents A and B, we can express the action performed by A with an decision function [6,18]. The resulting action is used to extend the current negotiation thread between the two agents.…”

A System for Analysis of Multi-Issue Negotiation

“…The effect of time on the negotiation equilibrium is the main feature studied, from both a game-theoretic and empirical perspective. In earlier research [20] a slightly different model is proposed, but the focus of the research is still on time constraints and the effect of deadlines on the agents' strategies. This contrast with our model, where efficiency of the outcome and not time is the main issue studied.…”

The Influence of Information on Negotiation Equilibrium

“…[13] models uncertainty over the player's discounting factors. [4] analyses bargaining in which reservation prices are uncertain, [14] considers uncertainty over agent deadlines and [2] addresses uncertainty over both deadlines and reservation prices. These existing models study the strategic behavior of agents by considering a particular scenario in which there is uncertainty over a chosen negotiation parameter.…”

“…Firstly, the deadline by when the negotiation must be completed (which may be different for each party). Secondly, the time at which agreement is reached can effect the agents in different ways [2]. An agent can gain utility with time and have the incentive to reach a late agreement (within its deadline).…”