Unveiling the Secrets of Diplomacy: How AI Agents Collaborate, Communicate, and Strategize
Research
- Published
- Authors
-
Yoram Bachrach, János Kramár
Agents cooperate better by communicating and negotiating, and sanctioning broken promises helps keep them honest
Successful communication and cooperation have been crucial for helping societies advance throughout history. The closed environments of board games can serve as a sandbox for modelling and investigating interaction and communication – and we can learn a lot from playing them. In our recent paper, published today in Nature Communications, we show how artificial agents can use communication to better cooperate in the board game Diplomacy, a vibrant domain in artificial intelligence (AI) research, known for its focus on alliance building.
What is Diplomacy and why is it important?
Games such as chess, poker, Go, and many other strategic games have always been fertile ground for AI research. Diplomacy is a seven-player game of negotiation and alliance formation, where each player controls multiple units on a map of Europe partitioned into provinces. The heart of Diplomacy lies in the negotiation phase, where players try to agree on their next moves.
Computational approaches to Diplomacy have been researched since the 1980s. Our recent study focuses on how AI agents can coordinate their moves through communication and negotiation protocols. These agents have shown significant improvement over non-communicating agents in the game.
Our negotiation mechanisms have proven to enhance the performance of AI agents in Diplomacy by identifying high-quality agreements and joint plans of action. Game theory principles are leveraged to ensure fairness and cooperation among the agents.
…
Continue on discussing different scenarios, agents breaking agreements, encouraging honesty, and conclusions