Security Games with Protection Externalities
Authors: Jiarui Gan, Bo An, Yevgeniy Vorobeychik
AAAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental evaluations are provided in this section to examine the performance of algorithms in this paper. All LPs and MILPs are solved with CPLEX (version 12.4). All results are obtained on a machine with a 3.10GHz quad core CPU and 4.00GB memory. We run our algorithm on game instances randomly generated in the following way. For each target i, Rd i and Ra i are randomly chosen between 0 and 100, and P d i and P a i are randomly chosen between 100 and 0. |
| Researcher Affiliation | Academia | Jiarui Gan The Key Laboratory of Intelligent Information Processing, ICT, CAS University of Chinese Academy of Sciences Beijing 100190, China ganjr@ics.ict.ac.cn Bo An School of Computer Engineering Nanyang Technological University Singapore 639798 boan@ntu.edu.sg Yevgeniy Vorobeychik Electrical Engineering & Computer Science Vanderbilt University Nashville, TN 37235 yevgeniy.vorobeychik@vanderbilt.edu |
| Pseudocode | Yes | Algorithm 1: A greedy algorithm for a WMC with non-negative weights |
| Open Source Code | No | The paper provides a URL for an appendix (http://www.ntu.edu.sg/home/boan/papers/ AAAI15 Externality Appendix.pdf) but does not state that source code for the methodology is available there, nor does it provide any other link to source code. |
| Open Datasets | No | We run our algorithm on game instances randomly generated in the following way. For each target i, Rd i and Ra i are randomly chosen between 0 and 100, and P d i and P a i are randomly chosen between 100 and 0. We set all diagonal entries of the adjacency matrix to 1, and set other entries to 1 with probability ρ and to 0 with 1 ρ. |
| Dataset Splits | No | The paper describes generating game instances randomly but does not provide specific dataset split information (e.g., percentages, sample counts, or predefined splits) for training, validation, or testing. |
| Hardware Specification | Yes | All results are obtained on a machine with a 3.10GHz quad core CPU and 4.00GB memory. |
| Software Dependencies | Yes | All LPs and MILPs are solved with CPLEX (version 12.4). |
| Experiment Setup | Yes | We run our algorithm on game instances randomly generated in the following way. For each target i, Rd i and Ra i are randomly chosen between 0 and 100, and P d i and P a i are randomly chosen between 100 and 0. We set all diagonal entries of the adjacency matrix to 1, and set other entries to 1 with probability ρ and to 0 with 1 ρ. Different settings of ρ are specified in the experiments. All the experimental results are averaged over 50 samples. |