Abstract University of Alberta expert poker agent: A survey
Bibliographic record
Abstract
Games have always been a natural topic for Artificial Intelligence researchers to study and poker has proven to be a game that is both interesting and challenging. Part of the challenge of poker comes from the fact that it is a game of imperfect knowledge where multiple competing agents must deal with risk management, agent modeling, unreliable information and deception, much like decision-making applications in the real world. In order to produce an expert level poker intelligent agent, researchers from the University of Alberta identified five characteristics of an expert poker player that must be modeled in the agent. Of these five characteristics, the characteristic that has been shown to improve quality of the play the most is opponent modeling. Initially, the researchers built their opponent models using a weight table and opponent action frequencies. This proved to be an improvement over previous versions with no opponent modeling, but required domain specific knowledge and proved to be difficult to maintain. In an effort to fix the problems with their previous opponent models, researchers began using neural networks and a training technique called backpropagation to accurately predict an opponent’s actions. After running the neural network training system on game played previously, the use of neural networks proved to be more accurate than previous attempts at opponent modeling. 1.
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 | 0.003 | 0.005 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.010 | 0.004 |
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".