# Algorithmic Stability and Generalization of an Unsupervised Feature Selection Algorithm

arXiv: https://arxiv.org/abs/2010.09416 · HTML: https://mlsanity.com/paper/2010.09416/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Oct 2020.

## Built on by (most cited first, 6 of 6)

- [Algorithmic Stability of Heavy-Tailed Stochastic Gradient Descent on Least Squares](https://mlsanity.com/paper/2206.01274/) (2022)
- [Composite Feature Selection using Deep Ensembles](https://mlsanity.com/paper/2211.00631/) (2022)
- [Explainable Censored Learning: Finding Critical Features with Long Term Prognostic Values for Survival Prediction](https://mlsanity.com/paper/2209.15450/) (2022)
- [Stability beyond Bounded Differences: Sharp Generalization Bounds under Finite Lp Moments](https://mlsanity.com/paper/2606.06855/) (2026)
- [Alcohol Intake Differentiates AD and LATE: A Telltale Lifestyle from Two Large-Scale Datasets](https://mlsanity.com/paper/2209.05438/) (2022)
- [i-IF-Learn: Iterative Feature Selection and Unsupervised Learning for High-Dimensional Complex Data](https://mlsanity.com/paper/2603.24025/) (2026)

## Abstract (arXiv)

Feature selection, as a vital dimension reduction technique, reduces data dimension by identifying an essential subset of input features, which can facilitate interpretable insights into learning and inference processes. Algorithmic stability is a key characteristic of an algorithm regarding its sensitivity to perturbations of input samples. In this paper, we propose an innovative unsupervised feature selection algorithm attaining this stability with provable guarantees. The architecture of our algorithm consists of a feature scorer and a feature selector. The scorer trains a neural network (NN) to globally score all the features, and the selector adopts a dependent sub-NN to locally evaluate the representation abilities for selecting features. Further, we present algorithmic stability analysis and show that our algorithm has a performance guarantee via a generalization error bound. Extensive experimental results on real-world datasets demonstrate superior generalization performance of our proposed algorithm to strong baseline methods. Also, the properties revealed by our theoretical analysis and the stability of our algorithm-selected features are empirically confirmed.