Streaming Sparse Principal Component Analysis

Authors: Wenzhuo Yang, Huan Xu

ICML 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Numerical experiments on synthetic and realworld datasets demonstrate good empirical performance of the proposed algorithms. We investigate the performance of our algorithms on a variety of simulated and real-world datasets.
Researcher Affiliation Academia Wenzhuo Yang A0096049@NUS.EDU.SG Department of Mechanical Engineering, National University of Singapore, Singapore 117576. Huan Xu MPEXUH@NUS.EDU.SG Department of Mechanical Engineering, National University of Singapore, Singapore 117576.
Pseudocode Yes Algorithm 1 Row Truncation Operator. Algorithm 2 Streaming SPCA via Row Truncation. Algorithm 3 Streaming SPCA via Iterative Deflation. Algorithm 4 Streaming ECA via Row Truncation. Algorithm 5 Finding Initial Solution.
Open Source Code No The paper does not provide any explicit statement or link for the open-sourcing of the described methodology's code.
Open Datasets Yes We use two large datasets, the NIPS paper dataset and the NYTimes news articles dataset, both available from the UCI Machine Learning Repository (Bache & Lichman).
Dataset Splits No The paper describes generating synthetic data and using real-world datasets with parameters like block size (B) and total samples (n). However, it does not explicitly provide information on standard train/validation/test splits, percentages, or sample counts needed for data partitioning.
Hardware Specification Yes The experiments are conducted on a desktop PC with an i7 3.4GHz CPU and 4G memory.
Software Dependencies No All the algorithms mentioned below are implemented in Python. This statement mentions the programming language but does not specify a version number or any other software libraries with their respective versions.
Experiment Setup Yes Parameters B and γ in streaming sparse PCA are set to 300 and 500, respectively. In the following experiments, the samples are independently drawn from ECp(0, Σ, ξ). Here, Σ is constructed according to Σ = AA + Ip where A is generated following the first scheme described above, and ξ follows the chi-distribution with degree of freedom p or the F-distribution with degrees of freedom p and 1.