Solving Patrolling Problems in the Internet Environment
Authors: Tomas Brazdil, Antonin Kucera, Vojtech Rehak
IJCAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The aim of our experiments is to demonstrate that the time needed to solve the constructed system of equations is low (using Maple), and also to show that the protection achieved by our strategies is substantially better than the protection achieved by naively constructed strategies. More concretely, we consider a surveillance system with three types of targets... |
| Researcher Affiliation | Academia | Tom aˇs Br azdil, Anton ın Kuˇcera and Vojtˇech ˇReh ak Faculty of Informatics, Masaryk University, Brno, Czech Republic |
| Pseudocode | No | The paper describes the 'Strategy Synthesis Algorithm' verbally in Section 5, explaining its decomposition and assignment procedures, but does not present a formal pseudocode block or algorithm listing. |
| Open Source Code | No | The paper refers to an arXiv preprint for the full version of the paper [Br azdil et al., 2018], but does not explicitly state that source code for the methodology is provided or offer a direct link to a code repository. |
| Open Datasets | No | The paper describes the parameters of the simulated environment ('surveillance system with three types of targets, where processing one image takes 0.1 sec... and the target values are $100000, $130000, and $400000, respectively') and how instances were generated ('the number of targets of each type is 7000000 x, 500000 x, and 300000 x, respectively'), but it does not specify a publicly available dataset, nor does it provide access information for the generated instances. |
| Dataset Splits | No | The paper discusses 'game structures' and 'instances' for analysis, but it does not specify any training, validation, or testing dataset splits, nor does it refer to predefined splits from established benchmarks. |
| Hardware Specification | No | The paper mentions that computations were performed 'on an average PC' but does not provide specific details on the hardware specifications (e.g., CPU model, GPU, memory). |
| Software Dependencies | No | The paper mentions the use of 'Maple' for solving systems of equations, but does not provide a specific version number for Maple or any other software dependencies. |
| Experiment Setup | Yes | More concretely, we consider a surveillance system with three types of targets, where processing one image takes 0.1 sec, an on-going intrusion is detected with probability 0.7, the time needed to complete an intrusion is either 20 secs, 2 mins, or 15 mins, and the target values are $100000, $130000, and $400000, respectively. We analyze large instances where the number of targets of each type is 7000000 x, 500000 x, and 300000 x, respectively, where the x ranges from 1 to 3 with a step 0.01. Hence, the total number of analyzed game structures is 200 and the largest one has more than 23 million vertices. The number of patrollers is set to 6000. |