Unsupervised Feature Selection Through Group Discovery
This work solves the problem of improving unsupervised feature selection for high-dimensional learning tasks where labels are unavailable, offering a novel approach that is not incremental but introduces a new method for a known bottleneck.
The paper tackled the problem of unsupervised feature selection by addressing the limitation of evaluating features in isolation, proposing GroupFS to jointly discover latent feature groups and select informative ones, resulting in consistent outperformance of state-of-the-art methods across nine benchmarks in clustering tasks.
Unsupervised feature selection (FS) is essential for high-dimensional learning tasks where labels are not available. It helps reduce noise, improve generalization, and enhance interpretability. However, most existing unsupervised FS methods evaluate features in isolation, even though informative signals often emerge from groups of related features. For example, adjacent pixels, functionally connected brain regions, or correlated financial indicators tend to act together, making independent evaluation suboptimal. Although some methods attempt to capture group structure, they typically rely on predefined partitions or label supervision, limiting their applicability. We propose GroupFS, an end-to-end, fully differentiable framework that jointly discovers latent feature groups and selects the most informative groups among them, without relying on fixed a priori groups or label supervision. GroupFS enforces Laplacian smoothness on both feature and sample graphs and applies a group sparsity regularizer to learn a compact, structured representation. Across nine benchmarks spanning images, tabular data, and biological datasets, GroupFS consistently outperforms state-of-the-art unsupervised FS in clustering and selects groups of features that align with meaningful patterns.