Learning and Using Hand Abstraction Values for Parameterized Poker Squares
Authors: Todd Neller, Colin Messinger, Zuozhi Yang
AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We describe the experimental development of an AI player that adapts to different point systems for Parameterized Poker Squares. After introducing the game and research competition challenge, we describe our static board evaluation utilizing learned evaluations of abstract partial Poker hands. Next, we evaluate various time management strategies and search algorithms. Finally, we show experimentally which of our design decisions most significantly accounted for observed performance. |
| Researcher Affiliation | Academia | Todd W. Neller and Colin M. Messinger and Zuozhi Yang Gettysburg College tneller@gettysburg.edu |
| Pseudocode | No | The paper does not contain any clearly labeled pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any concrete access information (e.g., repository link, explicit statement of code release) for the source code of the methodology described. |
| Open Datasets | No | For each random point system (the Ameritish point system, Random point system, Hypercorner point system) we generated a sample of 500 systems and measured the average greedy-play performance of 2000 games for each player and system. For fixed point systems, we collected average performance of 2000 games for each player and system. The paper describes generating custom point systems for the experiment, but does not provide access information (link, DOI, citation) for a publicly available or open dataset. |
| Dataset Splits | No | The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits) needed to reproduce data partitioning for typical machine learning experiments. While it discusses 'learning' and 'simulation' iterations, it doesn't describe data splits for a dataset. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory, or detailed computer specifications) used for running its experiments. |
| Software Dependencies | No | The paper does not provide specific ancillary software details (e.g., library or solver names with version numbers like Python 3.8, CPLEX 12.4) needed to replicate the experiment. |
| Experiment Setup | Yes | In our initial application of ϵ-greedy play, ϵ = 0.1 with geometric ϵ-decay δ = 0.999975 per simulated game iteration. However, we empirically observed that if we significantly raise the initial value of ϵ to 0.5, increasing initial exploration, the player has a better performance. We experimented with 3 players, all of which used ϵ-decay δ = 0.999975. The first used an initial epsilon ϵ0 = 0.1, whereas the second and third used ϵ0 = 0.5. |