# Scalable unsupervised feature selection via weight stability

arXiv: https://arxiv.org/abs/2506.06114 · HTML: https://mlsanity.com/paper/2506.06114/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 44 references that are in the map.
- First posted on arXiv 6 Jun 2025.

## Builds on (most cited first, 6 of 6)

- [TabNet: Attentive Interpretable Tabular Learning](https://mlsanity.com/paper/1908.07442/) (2019)
- [Infinite Feature Selection: A Graph-based Feature Filtering Approach](https://mlsanity.com/paper/2006.08184/) (2020)
- [Deep Unsupervised Feature Selection by Discarding Nuisance and Correlated Features](https://mlsanity.com/paper/2110.05306/) (2021)
- [A Survey on Soft Subspace Clustering](https://mlsanity.com/paper/1409.5616/) (2014)
- [Unsupervised Feature Selection based on Adaptive Similarity Learning and Subspace Clustering](https://mlsanity.com/paper/1912.05458/) (2019)
- [Adaptive Explicit Kernel Minkowski Weighted K-means](https://mlsanity.com/paper/2012.02805/) (2020)

## Abstract (arXiv)

Unsupervised feature selection is critical for improving clustering performance in high-dimensional data, where irrelevant features can obscure meaningful structure. In this work, we propose the Minkowski weighted k-means++, a novel initialisation strategy for the Minkowski Weighted k-means. Our initialisation selects centroids probabilistically using feature relevance estimates derived from the data itself. Building on this, we propose two new feature selection algorithms, FS-MWK++, which aggregates feature weights across a range of Minkowski exponents identifying stable and informative features, and SFS-MWK++, a scalable variant based on subsampling. We support our approach with a theoretical analysis, demonstrating that, under explicit assumptions on noise features and cluster structure, relevant features are assigned consistently higher weights than noise features across a range of Minkowski exponents. Our software can be found at https://github.com/xzhang4-ops1/FSMWK.