A Deterministic Analysis of Noisy Sparse Subspace Clustering for Dimensionality-reduced Data

Authors: Yining Wang, Yu-Xiang Wang, Aarti Singh

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

Reproducibility Variable Result LLM Response
Research Type Experimental In this section we present numerical results that validate our theoretical findings and compare Lasso SSC with TSC (Heckel & Bolcskei, 2013) and LRR (Liu et al., 2013). The Lasso SSC algorithm is implemented using augmented Lagrangian method (ALM) when the regularization coefficient λ is fixed and known. We also implement Lasso SSC using a solution path algorithm (Tibshirani & Taylor, 2011) to tune λ separately for each data point. The LRR implementation is obtained from (Liu, 2013). Random Gaussian projection is used for all experiments. All algorithms are implemented in Matlab.
Researcher Affiliation Academia Yining Wang YININGWA@CS.CMU.EDU Yu-Xiang Wang YUXIANGW@CS.CMU.EDU Aarti Singh AARTI@CS.CMU.EDU Machine Learning Department, Carnegie Mellon University, Pittsburgh, PA 15213, USA
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide concrete access to source code for the methodology described. It mentions 'The LRR implementation is obtained from (Liu, 2013)' but this refers to a third-party implementation, not their own.
Open Datasets Yes We start by evaluating the performance of Lasso SSC with random Gaussian projection on the extended Yale B face recognition dataset (Lee et al., 2005)... We evaluate the performance of Lasso SSC with random projection for motion trajectory segmentation on the Hopkins-155 dataset (Tron & Vidal, 2007).
Dataset Splits No The paper does not provide specific dataset split information (e.g., percentages, sample counts, or detailed methodology for training/validation/test splits).
Hardware Specification No The paper does not provide specific hardware details (exact GPU/CPU models, memory amounts, or detailed computer specifications) used for running its experiments.
Software Dependencies No All algorithms are implemented in Matlab. The paper does not specify a version number for Matlab or any other software dependencies.
Experiment Setup Yes The Lasso SSC algorithm is implemented using augmented Lagrangian method (ALM) when the regularization coefficient λ is fixed and known. We also implement Lasso SSC using a solution path algorithm (Tibshirani & Taylor, 2011) to tune λ separately for each data point. Random Gaussian projection is used for all experiments.