We demonstrate GNOME (Generating Novelty in Open-world Multi-agent Environments),
an experimental platform that is designed to test the effectiveness of multi-agent AI systems when faced with novelty.
GNOME separates the development of AI gameplaying agents with the simulator, allowing unanticipated novelty (in essence,
novelty that is not subject to model-selection bias). Through the demonstration, we hope to foster an open discussion
on AI robustness and the nature of novelty in real-world environments.
The GNOME simulator is a development platform for multi-agent boardgames of strategy.
Features of the GNOME simulator are :
Modular and Robust
Object Oriented Programming
Decision Making Interface
We demonstrate its facilities using the classic monopoly boardgame.
Users will have a range of agents to choose from to plug into the four players that play the monopoly game.
These agents range from simple and complex heuristic agents to machine learning based agents.
Users can also choose the novelty scenario they wish to inject into the gameboard.
The game with novelty gets played out and users will get an opportunity to see the difference in player performance with
different agent combinations in the presence of novelties.