Automated negotiation has been an active research topic. Many previous works try to model opponents’ behaviours using some mathematical functions. However, this approach turns out not to be very effective due to inaccessibility of information and dynamics of the negotiation environment. In this paper,we introduce an adaptive strategy for automated negotiation.We propose that only the concession degree of the opponent is estimated in the negotiation.We also use an adaptive nonexploitation point on the timeline of negotiation for stopping making concession to the opponent. Finally, we suggest that the bid that we propose should be as close to the Pareto Frontier of the bids as possible. The agent CUHKAgent2015 that is based on this strategy showed good performance in ANAC 2015.
CITATION STYLE
Ng, C. W., Leung, H. T., & Leung, H. F. (2017). CUHKAgent2015: An adaptive negotiation strategy in multilateral scenario. Studies in Computational Intelligence, 674, 243–250. https://doi.org/10.1007/978-3-319-51563-2_20
Mendeley helps you to discover research relevant for your work.