Nonnegative Sparse PCA with Provable Guarantees

Authors: Megasthenis Asteris, Dimitris Papailiopoulos, Alexandros Dimakis

ICML 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We test our scheme on several data sets, showing that it matches or outperforms the previous state of the art. and 7. Experimental Evaluation We empirically evaluate the performance of our algorithm on various datasets and compare it to the EM algorithm
Researcher Affiliation Academia Megasthenis Asteris MEGAS@UTEXAS.EDU Dimitris S. Papailiopoulos DIMITRIS@UTEXAS.EDU Alexandros G. Dimakis DIMAKIS@AUSTIN.UTEXAS.EDU Department of Electrical and Computer Engineering, The University of Texas at Austin, TX, USA
Pseudocode Yes Algorithm 1 Spannogram Nonnegative Sparse PCA
Open Source Code No Matlab implementation available by the author. - This statement is not concrete access to a repository or an explicit release statement.
Open Datasets Yes CBCL Face Dataset (Sung, 1996), Leukemia Dataset (Armstrong et al., 2001), and Low Resolution Spectrometer (LRS) dataset, available in (Bache & Lichman, 2013)
Dataset Splits No The paper references datasets but does not provide specific details on training, validation, or test splits, such as percentages, sample counts, or explicit splitting methodologies.
Hardware Specification No The paper does not provide specific hardware details (e.g., exact GPU/CPU models, processor types, or memory amounts) used for running its experiments.
Software Dependencies No The paper mentions 'Matlab implementation' but does not specify any software names with version numbers for dependencies.
Experiment Setup Yes Alg. 3 for d = 3 and ϵ = 0.1, and the EM algorithm exhibit nearly identical performance. and an appropriate sparsity penalty β was determined via binary search for each target sparsity k.