Computational Modelling Group

Simulating Multi-Agent Negotiation with Agents that use Incomplete Information

Research Team
Darius Pepe Falahat
Investigators
Enrico Gerding

The program models bilateral negotiation (using the alternating-offers protocol) between a buyer and a seller, who have incomplete information about each other and have to guess their opponent's true parameters. The model was extended beyond the initial ideas from the literature by letting the agents adapt (based on past experiences) to improve their beliefs about each other. A further extension was letting agents use tit-for-tat (TFT) strategies, and experimenting with how these interacted with the original time dependent strategies.

Categories

Algorithms and computational methods: Agent-Based Negotiation, Agents

Programming languages and libraries: C

Transdisciplinary tags: Complex Systems