Accelerated Best-First Search With Upper-Bound Computation for Submodular Function Maximization

Authors: Shinsaku Sakaue, Masakazu Ishihata

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments show that our accelerated BFS is far more efficient in terms of both time and space complexities than existing methods.
Researcher Affiliation Collaboration Shinsaku Sakaue NTT Communication Science Laboratories sakaue.shinsaku@lab.ntt.co.jp Masakazu Ishihata Hokkaido University ishihata.masakazu@ist.hokudai.ac.jp
Pseudocode Yes Algorithm 1 BFSTC(U, g( ), c U, B, α); Algorithm 2 Greedy SK(V, g S( ), c V , BS); Algorithm 3 Greedy Add(V, g S( ), c V , BS)
Open Source Code No The paper does not provide any explicit statements about releasing source code for the described methodology, nor does it include links to a code repository.
Open Datasets Yes COV: The COV instance uses a dataset on messages exchanged by 899 users on 522 topics (Opsahl 2013). https://toreopsahl.com/datasets/#online forum network. LOC: The LOC instance uses a dataset on 473 subway stations in New York City (NYC) (Roest and Mashariki 2015). https://data.cityofnewyork.us/Transportation/Subway-Stations/arq3-7z49/data. INF: The INF instance uses the Movie Lens 100K dataset (Harper and Konstan 2015). https://grouplens.org/datasets/movielens/.
Dataset Splits No The paper describes the datasets used for experiments but does not provide specific details on train/validation/test splits (e.g., percentages or sample counts) for reproducibility.
Hardware Specification Yes All experiments were conducted on a 64-bit Cent7.3.1611 machine with Xeon 5E-2697 v3 2.6 GHz CPUs and 256 GB of RAM.
Software Dependencies No The paper mentions the operating system '64-bit Cent7.3.1611' but does not specify any programming languages, libraries, or other software dependencies with version numbers used for the experiments.
Experiment Setup Yes In all instances, we set |U| = 100 and B = 1. The costs cv (v U) are drawn randomly from a uniform distribution on [0, 1]... In COV instances, wi are drawn from u[0,1], and each v U randomly covers each w W with probability 0.3.