Decoding the Chameleon Game
The Chameleon game is a challenging word association activity where players are given a secret word and must respond with words relevant to that secret word. It requires strategic thinking and deduction. The Chameleon must cleverly guess the secret keyword in this game while avoiding suspicion. Our research aims to create an advanced artificial intelligence (AI) model that can play the Chameleon game from both perspectives: as the Chameleon and as a Human. This AI is designed to guess secret keywords based on the information provided by the players, choose the best strategies to avoid detection as the Chameleon, identify and vote for the most suspicious players when acting as a Human, and learn from previous games to improve its performance.
Authors
- Dat Tien Nguyen (ORCID: https://orcid.org/0000-0002-6029-6849)
- Brian M. Howard (ORCID: https://orcid.org/0000-0001-9134-0817)
- Hieu Minh Tran (ORCID: https://orcid.org/0000-0002-5552-2880)
- Tri Vi Dang
- Sutthirut Charoenphon
Institutions
- DePauw University (US)
- Purdue University West Lafayette (US)
Publication Details
- Journal
- AI Matters
- Published
- 2025-10-01
- DOI
- https://doi.org/10.1145/3774399.3774403
- Primary Topic
- Artificial Intelligence in Games
- Type
- article
- Field-Weighted Citation Impact
- 0.00