Affiliations: [a] Computer Science Department, University of Massachusetts, Amherst, MA 01003, USA | [b] Department of Computer Science, Hong Kong Baptist University, Kowloon Tong, Hong Kong | [c] School of Computer Engineering, Nanyang Technological University, Singapore | [d] School of Electrical & Electronic Engineering, Nanyang Technological University, Singapore | [x] Hong Kong Baptist University, Kowloon, Hong Kong, China | [y] University of Wollongong, Wollongong, NSW 2522, Australia | [z] Nagoya Institute of Technology, Gokiso, Showa, Nagoya 466-8555, Japan
Abstract: In dynamic and complex negotiation environments, a negotiation agent can participate or quit negotiation at any time and can potentially reach an agreement with more than one trading partner as the result of the existence of dynamic outside options. Thus, it's important for a negotiation agent to make a decision on when to complete negotiation given its trading partners' current proposals and market dynamics. Rather than explicitly modeling all the trading partners, this paper presents a novel decision making strategy based on a tractable Markov chain model of negotiation process. An agent can use this model to determine whether to accept the best proposal of its trading partners or let negotiation proceed forward during each round of negotiation. Experimental results suggest that the proposed strategy achieved more favorable negotiation outcomes as compared with the general strategy.
