Coarse Correlation in Extensive-Form Games
Authors: Gabriele Farina, Tommaso Bianchi, Tuomas Sandholm1934-1941
AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments We experimentally compared NFCCE, EFCCE, and EFCE in terms of maximum social welfare and run time. In our experiments, we used instances from three different two-player games with no chance moves: Sheriff (Farina et al. 2019), Battleship (Farina et al. 2019), and Goofspiel (Ross 1971). |
| Researcher Affiliation | Collaboration | Gabriele Farina,1 Tommaso Bianchi,2 Tuomas Sandholm1,3,4,5 1Computer Science Department, Carnegie Mellon University, 2DEIB, Politecnico di Milano 3Strategic Machine, Inc., 4Strategy Robot, Inc., 5Optimized Markets, Inc. |
| Pseudocode | No | The paper does not include any explicit pseudocode or algorithm blocks. |
| Open Source Code | No | The paper provides a link to its arXiv version for supplemental material, but does not state that source code for the described methodology is available, nor does it provide a direct link to a code repository. |
| Open Datasets | No | The paper names specific games ("Sheriff", "Battleship", "Goofspiel") and references their sources, but does not provide concrete access information (links, DOIs, repositories) for the specific game instances or "datasets" used in their experiments. |
| Dataset Splits | No | The paper discusses game instances for problem-solving rather than datasets with train/validation/test splits. Therefore, it does not provide specific dataset split information. |
| Hardware Specification | Yes | All experiments were run on a 64-core machine with 512 GB of RAM. |
| Software Dependencies | Yes | We used Gurobi 8.1.1 (Gurobi Optimization 2019) to solve the linear programs for NFCCE (8), EFCCE (12), and EFCE (Equation 16 in the Supplemental Material). |
| Experiment Setup | Yes | We used the barrier algorithm without crossover, and we let Gurobi automatically determine the recommended number of threads for execution. |