M@sterMind : a diplomacy game negotiating agent

• Main Author: Ng, Jonathan Sheng Guo
• Other Authors: Bo An
• Format: Final Year Project
• Language: English
• Published: 2018
• Subjects: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
• Online Access: http://hdl.handle.net/10356/76131
• Institution: Nanyang Technological University

Game playing has always been a rich environment for researchers and developers of intelligent agents using innovate concepts. Diplomacy is one of such games that is being used to conduct in that field of research. The ANAC 2018 Diplomacy Challenge is a competition that encourage research in intelligent agents. This paper introduces M@sterMind a Diplomacy agent with a negotiating algorithm developed in accordance with the regulations of the ANAC 2018 Diplomacy Challenge. The negotiating algorithm is split into 2 components, the deal consideration and the deal proposal. Deal consideration considers all incoming deal. The Agent will carefully evaluate each component of the deal and will return an acceptance probability. The average of all acceptance probability will be considered and the Agent will take action accordingly.Deal proposal propose deals to other agents in the game. The Agent will ensure that these deals are Pareto Optimal before sending them. The purpose of these deals is to deter invaders and to prevent any possible conflict with an opposing agent. The Agent also propose Demilitarized Zones to protect it provinces from threats in the long-term. These negotiations are bilateral as adding more parties to a negotiation will increase the number of issues that require resolution in the negotiation. During testing, the Agent has shown performance comparable to agents that have obtained some degree of success in the Diplomacy Challenge.There needs to be more testing on the Agent to conclude its effectiveness. This is due to the lack of available resources which is attributed to the newness of the Challenge.