Quasi-Perfect Stackelberg Equilibrium

Authors: Alberto Marchesi, Gabriele Farina, Christian Kroer, Nicola Gatti, Tuomas Sandholm2117-2124

AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments show that our algorithm can be used to find an approximate quasi-perfect Stackelberg equilibrium in games with thousands of nodes.
Researcher Affiliation Academia 1 Politecnico di Milano, Piazza Leonardo da Vinci 32, I-20133, Milan, Italy 2 Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213, USA
Pseudocode No The paper describes an algorithm and provides mathematical formulations (LP), but it does not include a clearly labeled 'Pseudocode' or 'Algorithm' block.
Open Source Code No The paper mentions using open-source tools like GLPK, but it does not state that the authors' own implementation code is open-source or provide a link to it.
Open Datasets No The paper conducts experiments on two common benchmark Extensive Form Games (EFGs), a search game and Goofspiel. However, it does not refer to these as 'datasets' in the traditional sense for which public access information (links, DOIs, specific citations to data repositories) is provided.
Dataset Splits No The paper does not explicitly specify training, validation, or test dataset splits. The experiments are conducted on game instances rather than partitioned datasets.
Hardware Specification Yes It was run on a single thread on a 2.3 GHz Intel Xeon processor.
Software Dependencies Yes The LP solver we used is GLPK 4.63 (GLPK 2017)... Our code and GLPK use the GNU GMP library... The code, written in the C++14 language, was compiled with the g++ 7.2.0 compiler.
Experiment Setup No The paper describes the setup of the game instances used for experiments but does not provide specific hyperparameter values or detailed system-level training settings for the algorithm itself.