Constant Approximation for Individual Preference Stable Clustering

Authors: Anders Aamand, Justin Chen, Allen Liu, Sandeep Silwal, Pattara Sukprasert, Ali Vakilian, Fred Zhang

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

Reproducibility Variable Result LLM Response
Research Type Experimental We experimentally evaluate our O(1)-IP stable clustering algorithm against k-means++, which is the empirically best-known algorithm in [1]. We also compare kmeans++ with our optimal algorithm for Min-IP stability. We run experiments on the Adult data set
Researcher Affiliation Collaboration Anders Aamand MIT aamand@mit.edu Justin Y. Chen MIT justc@mit.edu Allen Li MIT cliu568@mit.edu Sandeep Silwal MIT silwal@mit.edu Pattara Sukprasert Databricks pat.sukprasert@databricks.com Ali Vakilian TTIC vakilian@ttic.edu Fred Zhang UC Berkeley z0@berkeley.edu
Pseudocode Yes Algorithm 1 BALL-CARVING
Open Source Code No The paper does not provide a direct link to its own open-source code or explicitly state that the code for its methods is publicly available. It only mentions using the 'k-means++ implementation of Scikit-learn package [21]'.
Open Datasets Yes We run experiments on the Adult data set2 https://archive.ics.uci.edu/ml/datasets/adult; see [18]. For IP stability, we also use four more datasets from UCI ML repositoriy [11]
Dataset Splits No The paper mentions using various datasets but does not provide specific details on the training, validation, or test dataset splits (e.g., percentages, sample counts, or references to predefined standard splits).
Hardware Specification No The paper does not provide specific details about the hardware used for running its experiments (e.g., GPU models, CPU models, or memory specifications).
Software Dependencies No The paper mentions 'Python 3' and 'Scikit-learn package [21]' but does not provide specific version numbers for these software dependencies, which are necessary for reproducible descriptions.
Experiment Setup Yes We use the k-means++ implementation of Scikit-learn package [21]... we set the parameter n_init=1 and then compute the average of many different runs.