Coordinate-wise Power Method

Authors: Qi Lei, Kai Zhong, Inderjit S. Dhillon

NeurIPS 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on both synthetic and real data show that our methods achieve up to 23 times speedup over the basic power method.
Researcher Affiliation Academia Qi Lei 1 Kai Zhong 1 Inderjit S. Dhillon 1,2 1 Institute for Computational Engineering & Sciences 2 Department of Computer Science University of Texas at Austin {leiqi, zhongkai}@ices.utexas.edu, inderjit@cs.utexas.edu
Pseudocode Yes Algorithm 1 Coordinate-wise Power Method
Open Source Code No The paper does not provide an explicit statement about the release of its source code or a link to a code repository.
Open Datasets Yes We use the following real datasets: 1) com-Orkut: Orkut online social network 2) soc-Live Journal: On-line community for maintaining journals, individual and group blogs 3) soc-Pokec: Pokec, most popular on-line social network in Slovakia 4) web-Stanford: Pages from Stanford University (stanford.edu) and hyperlinks between them 5) ego-Gplus (Google+): Social circles from Google+ 6) ego-Twitter: Social circles from Twitter. The Twitter dataset is further referenced by: [10] Haewoon Kwak, Changhyun Lee, Hosung Park, and Sue Moon. What is Twitter, a social network or a news media? Proceedings of the 19th international conference on World wide web, pages 591 600, 2010.
Dataset Splits No The paper does not explicitly provide specific dataset split information (e.g., percentages, sample counts, or detailed splitting methodology) for training, validation, or testing.
Hardware Specification Yes All the experiments were executed on Intel(R) Xeon(R) E5430 machine with 16G RAM and Linux OS.
Software Dependencies No The paper states, 'We implement all the five algorithms in C++ with Eigen library.' However, it does not provide specific version numbers for the Eigen library or any other software dependencies, making the setup not fully reproducible.
Experiment Setup Yes For each algorithm, we start from the same random vector, and set stopping condition to be cos^-1 (x, v_1) < , = 10^-6.