Safe and Nested Subgame Solving for Imperfect-Information Games

Authors: Noam Brown, Tuomas Sandholm

NeurIPS 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our experiments were conducted on heads-up no-limit Texas hold em, as well as two smaller-scale poker games we call No-Limit Flop Hold em (NLFH) and No-Limit Turn Hold em (NLTH). Table 1 shows the performance of each technique when using 30,000 buckets (20,000 for NLTH).
Researcher Affiliation Academia Noam Brown Computer Science Department Carnegie Mellon University Pittsburgh, PA 15217 noamb@cs.cmu.edu Tuomas Sandholm Computer Science Department Carnegie Mellon University Pittsburgh, PA 15217 sandholm@cs.cmu.edu
Pseudocode No The paper describes methodologies in prose but does not include structured pseudocode or algorithm blocks.
Open Source Code No The paper mentions Libratus was developed using these techniques but does not provide a statement or link for open-sourcing the code described in the paper.
Open Datasets No Our experiments were conducted on heads-up no-limit Texas hold em, as well as two smaller-scale poker games we call No-Limit Flop Hold em (NLFH) and No-Limit Turn Hold em (NLTH). The paper uses game environments rather than distinct datasets with explicit public availability information.
Dataset Splits No The paper describes experiments in game environments and discusses information abstraction, but does not specify train/validation/test dataset splits needed to reproduce the experiment.
Hardware Specification No The paper mentions using 'XSEDE computing resources provided by the Pittsburgh Supercomputing Center' but does not specify particular GPU models, CPU types, or detailed hardware specifications.
Software Dependencies No The paper states 'For equilibrium finding, we used CFR+ [30]' but does not provide specific version numbers for CFR+ or any other software dependencies.
Experiment Setup No The paper details game parameters such as bet sizes and abstraction bucket counts, but does not provide specific hyperparameters like learning rates, batch sizes, or optimizer settings typically found in an experimental setup section.