A Fuzzy Constraint-Directed Autonomous Learning to Support Agent Negotiation

Abstract

This work presents a general framework of agent negotiation with autonomous learning via fuzzy constraint-directed approach. The fuzzy constraint-directed approach involves the fuzzy probability constraint where each fuzzy constraint has a certain probability, and the fuzzy instance reasoning where each instance is represented as a primitive fuzzy constraint network. The proposed approach via fuzzy probability constraint can not only cluster the opponent's information in negotiation process as proximate regularities to increase the efficiency on the convergence of behavior patterns, but also eliminate the bulk of false hypotheses or beliefs to improves the effectiveness on beliefs learning. By using fuzzy instance method, our approach can reuse the prior opponent knowledge to speed up problem-solving, and reason the proximate regularities to acquire desirable results on predicting opponent behavior. Besides, the proposed interaction method enables the agent to make a concession dynamically based on expected objectives. Moreover, experimental results suggest that the proposed framework allowed an agent to achieve a higher reward, fairer deal, or less cost of negotiation.

Multi-issue Agent Negotiation Based on Fairness

“…This paper references the literature [7], makes an improvement on utility function, bring forward the satisfaction evaluation function based on satisfaction of both sides:…”

“…Agent auto-negotiation model based on the history of interaction [4], [5] can gain knowledge from opponent's interactive history prior to the negotiation, form the initial belief in negotiation, estimate opponent's strategy more effectively, shorten the negotiation time and increase the success rate of negotiation, but they hardly consider fairness. On the other hand, fuzzy logic-based Agent auto-negotiation model [6], [7] can carry through bilateral negotiation with less initial faith, but the negotiation time is often too long. This paper combining the advantages of negotiation theories [4], [5], [6], [7] and learning the opponent's offer to further study the opponent's strategy, proposed a optimal multi-issue bilateral negotiation framework based on the fairness.…”

“…On the other hand, fuzzy logic-based Agent auto-negotiation model [6], [7] can carry through bilateral negotiation with less initial faith, but the negotiation time is often too long. This paper combining the advantages of negotiation theories [4], [5], [6], [7] and learning the opponent's offer to further study the opponent's strategy, proposed a optimal multi-issue bilateral negotiation framework based on the fairness.…”