Replicable Clustering

Authors: Hossein Esfandiari, Amin Karbasi, Vahab Mirrokni, Grigoris Velegkas, Felix Zhou

NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We propose such algorithms for the statistical k-medians, statistical k-means, and statistical k-centers problems by utilizing approximation routines for their combinatorial counterparts in a black-box manner. In addition, we provide experiments on synthetic distributions in 2D using the k-means++ implementation from sklearn as a black-box that validate our theoretical results.
Researcher Affiliation Collaboration Hossein Esfandiari Google Research esfandiari@google.com Amin Karbasi Yale University, Google Research amin.karbasi@yale.edu Vahab Mirrokni Google Research mirrokni@google.com Grigoris Velegkas Yale University grigoris.velegkas@yale.edu Felix Zhou Yale University felix.zhou@yale.edu
Pseudocode Yes Algorithm 4.1 Replicable Quad Tree; Algorithm D.1 Replicable k-Means with ε-Cover; Algorithm D.2 Replicable Heavy Hitters; Algorithm D.3 Replicable Coreset; Algorithm D.4 Replicable Rounding; Algorithm F.1 Oracle with Grid; Algorithm F.2 Replicable Active Cells
Open Source Code Yes https://anonymous.4open.science/r/replicable_clustering_experiments-E380
Open Datasets No The paper mentions "synthetic distributions in 2D using the k-means++ implementation from sklearn" and specifically refers to "two moons distribution" and "mixture of truncated Gaussian distributions." However, it does not provide any concrete access information (link, DOI, citation) to these synthetic datasets.
Dataset Splits No The paper conducts experiments on synthetic data (
Hardware Specification No The paper does not provide specific details about the hardware used for the experiments.
Software Dependencies No The paper mentions using "sklearn" and "k-means++" but does not specify version numbers for these software components.
Experiment Setup No The paper mentions using "k = 3" for k-means++ in the experiments. No other specific hyperparameters or training configurations are provided for the experimental setup.