Hybrid Multi-Agent AI/MCTS Systems for Complex Information-Imperfect Games
Bibliographic record
Abstract
This paper presents a comprehensive artificial intelligence system for imperfect information games, demonstrated through the complex card game “28”. 28 is a team-based trick-taking card game, featuring bidding, trump selection and dynamic gameplay. Our system introduces several contributions: (1) a hybrid decision-making framework that dynamically combines belief networks, Monte Carlo Tree Search (MCTS), and reinforcement learning; (2) an innovative point prediction approach leading to accurate bids; (3) an advanced belief network architecture for opponent modeling, predicting the trump that an opponent might set; and (4) an Information Set MCTS (ISMCTS) implementation that handles imperfect information scenarios. The system achieves significant performance improvements through multi-modal learning from 3873 MCTS-generated games, demonstrating the effectiveness of combining multiple AI paradigms for complex game environments. Our experimental results show that the hybrid approach outperforms individual methods by 15-25% in win rates, while the belief network achieves 70-80% accuracy in opponent hand prediction. The system’s modular architecture enables real-time decision-making while maintaining strategic depth, making it suitable for adaptation to other imperfect information games.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".