CLAMS: A System for Zero-Shot Model Selection for Clustering
Machine Learning
2024-07-17 v1 Artificial Intelligence
Abstract
We propose an AutoML system that enables model selection on clustering problems by leveraging optimal transport-based dataset similarity. Our objective is to establish a comprehensive AutoML pipeline for clustering problems and provide recommendations for selecting the most suitable algorithms, thus opening up a new area of AutoML beyond the traditional supervised learning settings. We compare our results against multiple clustering baselines and find that it outperforms all of them, hence demonstrating the utility of similarity-based automated model selection for solving clustering applications.
Cite
@article{arxiv.2407.11286,
title = {CLAMS: A System for Zero-Shot Model Selection for Clustering},
author = {Prabhant Singh and Pieter Gijsbers and Murat Onur Yildirim and Elif Ceren Gok and Joaquin Vanschoren},
journal= {arXiv preprint arXiv:2407.11286},
year = {2024}
}