Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Potential-Aware Imperfect-Recall Abstraction with Earth Mover’s Distance in Imperfect-Information Games
Authors: Sam Ganzfried, Tuomas Sandholm
AAAI 2014 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments on no-limit Texas Hold em show that our algorithm improves performance over the previously best approach. . . . In both experiments, our new approach led to a statistically significant improvement over the old approach. |
| Researcher Affiliation | Academia | Sam Ganzfried and Tuomas Sandholm Computer Science Department Carnegie Mellon University EMAIL |
| Pseudocode | Yes | Our abstraction algorithm, depicted in Algorithm 1, works as follows. . . . Our heuristic is given in Algorithm 2. |
| Open Source Code | No | The paper does not include an unambiguous statement or link indicating that the source code for the described methodology is publicly available. |
| Open Datasets | No | The paper evaluates an algorithm within the game of no-limit Texas Hold 'em and does not refer to a publicly available or open dataset with specific access information for training or evaluation. The 'data' mentioned for figures 1 and 2 is specific to those visualizations, not the core experimental setup. |
| Dataset Splits | No | The paper does not provide specific dataset split information (like percentages or sample counts for training, validation, and test sets) needed to reproduce data partitioning. |
| Hardware Specification | No | The paper mentions 'parallelizing each step using 64 cores' but does not provide specific details such as CPU/GPU models, processor types, or memory amounts used for the experiments. |
| Software Dependencies | No | The paper mentions algorithms and external references (e.g., 'k-means++', 'counterfactual regret minimization', 'multi-dimensional EMD algorithm') but does not list specific software components with version numbers required for reproduction. |
| Experiment Setup | Yes | In both experiments, we used 169, 5000, 5000, and 5000 card buckets respectively in the four betting rounds for both the new algorithm and the prior algorithm. . . . For both flop abstraction algorithms, we conducted 25 restarts using the k-means++ initialization procedure (Arthur and Vassilvitskii 2007), and selected the run that produced the lowest within-cluster sum of squares. . . . In each of the two experiments, we ran 20,000 duplicate matches between our new agent and the respective benchmark agent. |