Sampling for Approximate Maximum Search in Factorized Tensor
Authors: Zhi Lu, Yang Hu, Bing Zeng
IJCAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We provide theoretical analysis of the sampling algorithm and evaluate its performance on several real-world data sets. Experimental results indicate that the proposed approach is orders of magnitude faster than exhaustive computing. |
| Researcher Affiliation | Academia | School of Electronic Engineering University of Electronic Science and Technology of China zhilu@std.uestc.edu.cn, {yanghu,eezeng}@uestc.edu.cn |
| Pseudocode | Yes | Algorithm 1 The Corek Sampling Method; Algorithm 2 Finding top-t entries for multiple users |
| Open Source Code | No | The paper does not explicitly state that source code for the methodology is released or provide a link to a code repository. |
| Open Datasets | Yes | We evaluate our algorithms on five real-world data sets: Delicious Bookmarks1(Delicious), Last.FM2(Last FM), Movie Lens3+IMDb4/Rotten Tomatoes5(ML-S), a larger Movie Lens data set (ML-L) and dblp6. (Footnotes: 1http://www.delicious.com, 2http://www.lastfm.com, 3http://www.grouplens.org, 4http://www.imdb.com, 5http://www.rottentomatoes.com, 6http://dblp.uni-trier.de/db/) |
| Dataset Splits | No | The paper does not provide explicit details about training, validation, and test dataset splits, or cross-validation methodology. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used for running the experiments. |
| Software Dependencies | No | The paper mentions software like Bayesian Probabilistic Tensor Factorization (BPTF), Bayesian Personalized Ranking (BPR), and Tensor Toolbox, but does not provide specific version numbers for these or any other ancillary software dependencies. |
| Experiment Setup | Yes | R is chosen following previous works, i.e. 64 for the first four data sets and 50 for the last one. We experiment with k equals 1,2 and 3 respectively. We set t = s and focus on the effectiveness of the sampling stage firstly. For each s, we run 10 times and the average result is reported. We set t = s/10. |