Data Clustering by Laplacian Regularized L1-Graph

Authors: Yingzhen Yang, Zhangyang Wang, Jianchao Yang, Jiangping Wang, Shiyu Chang, Thomas Huang

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

Reproducibility Variable Result LLM Response
Research Type Experimental The experimental results on real data sets demonstrate the superiority of our algorithm compared to ℓ1-Graph and other competing clustering methods.
Researcher Affiliation Collaboration 1 Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA. 2 Adobe Research, San Jose, CA 95110, USA {yyang58, zwang119, jwang63, chang87, huang}@ifp.uiuc.edu, jiayang@adobe.com
Pseudocode Yes Algorithm 1 describes the learning procedure for LRℓ1-Graph. Algorithm 1 Learning Procedure for LRℓ1-Graph
Open Source Code No The paper does not provide concrete access to source code for the methodology described.
Open Datasets Yes The clustering results on several real data sets, i.e. UCI Wine, UCI Breast Tissue (BT) and ORL face database, are shown in Table 1 where the clustering performance is measured by Accuracy (AC) and the Normalized Mutual Information (NMI).
Dataset Splits No The paper mentions using real datasets (UCI Wine, UCI Breast Tissue, ORL face database) but does not specify details about training, validation, or test splits. It directly presents 'Clustering Results'.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory amounts) used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details with version numbers.
Experiment Setup Yes We use fixed empirical value λ = 0.1, γ = 30, M = 2 throughout the experiments, and tune λ between [0.1, 1] for ℓ1-Graph and SMCE.