Clustering in the Presence of Background Noise

Authors: Shai Ben-David, Nika Haghtalab

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

Reproducibility Variable Result LLM Response
Research Type Experimental In Section 8, we further demonstrate the gap between these two paradigms and use experiments to confirm our theoretical results.
Researcher Affiliation Academia David R. Cheriton School of Computer Science, University of Waterloo, Waterloo, ON N2L 3G1 CANADA; Computer Science 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.
Open Datasets No The paper uses synthetically generated data and does not provide access information for a publicly available dataset: 'For any k, we use n = 50000 data points on the unit square. 90% of the points come from k Gaussian distributions with centers selected uniformly at random and standard deviation = 1/n. Additionally, 10% uniform noise is introduced in the data.'
Dataset Splits No The paper describes the synthetic data generation but does not provide specific dataset split information (e.g., percentages, sample counts, or explicit validation splits).
Hardware Specification No No specific hardware details (GPU/CPU models, memory, or detailed computer specifications) used for running experiments were found.
Software Dependencies No The paper mentions algorithms like Lloyd but does not provide specific software names with version numbers (e.g., library or solver names with version numbers).
Experiment Setup Yes For δ-k-means, we adapt the Lloyd algorithm to calculate the clustering using a δ-truncated distance matrix in every iteration, where δ is set to 10.