Robust Principal Component Analysis with Adaptive Neighbors

Authors: Rui Zhang, Hanghang Tong

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental 5 Experiment Diverse experiments are conducted to evaluate the performance of our method. Firstly, the experimental settings are provided. Moreover, the experimental results on different tasks are recorded.
Researcher Affiliation Academia Rui Zhang Arizona State University Tempe, AZ, U.S.A. ruizhang8633@gmail.com Hanghang Tong University of Illinois at Urbana-Champaign Urbana, Illinois, U.S.A. htong@illinois.edu
Pseudocode Yes Algorithm 1: Algorithm for solving RWL-AN in (6)
Open Source Code No The paper does not provide any specific repository link or explicit statement about the public availability of the source code for the described methodology.
Open Datasets Yes Four benchmark face image datasets including AT&T [1], UMIST [3], FEI and FERET [12] are utilized in the experiment. Table 1 reports the information for the benchmark datasets.
Dataset Splits No The paper mentions running experiments '50 times with random initialization' and uses a 'noise rate' for the datasets, but it does not specify explicit training/validation/test splits or cross-validation setup for reproducing the data partitioning.
Hardware Specification Yes Note that all the experiments are implemented by MATLAB R2015b on Windows 7 PC with 3.20 GHz i5-3470 CPU and 16.0 GB main memory.
Software Dependencies Yes Note that all the experiments are implemented by MATLAB R2015b on Windows 7 PC...
Experiment Setup Yes The integer parameter k of our method is setted as [0.85N] (N is the total number of data points), such that 85% samples are assigned with non-zero weights. As for capped R2DPCA, ϵ is searched in the grid of {10, 20, ..., 50} and the best results are recorded.