aiqbench analysis
This paper introduces LIDFS, an unsupervised feature selection method that ranks features by their local intrinsic dimensionality, without labels or class counts. Its reported results, on 23 benchmark datasets, are competitive with supervised-style baselines and faster than its closest rival, and it notably beats several recent methods that fail to outperform a random baseline. For ML practitioners, the key questions are how well these benchmark gains transfer to real-world, noisy, or streaming data, and whether the O(d²n) cost holds at very high dimensionality.
Unsupervised Feature Selection Using Local Intrinsic Dimensionality -- Extended Version
arXiv:2610.40736v1 Announce Type: new Abstract: High dimensionality degrades both the efficiency and the effectiveness of learning algorithms, and in many applications no class labels are available to guide dimensionality reduction. This paper proposes LIDFS, an unsupervised feature selection method based on Local Intrinsic Dimensionality (LID). The method builds on the formal equivalence between LID and the indiscriminability of distance measures. The data matrix is transposed so that each feature is treated as a point, and the LID of each feature is estimated by maximum likelihood from the distances to its $k$ nearest neighboring features. Features with lower LID exhibit greater local discriminability and receive higher importance scores. A refinement step then retains one representative per feature neighborhood and demotes locally redundant features. The result is a complete ranking from which any number of features can be selected without rerunning the procedure. LIDFS does not use any information about labels or the number of classes, and runs in $\mathcal{O}(d^2 n)$ time for $n$ instances and $d$ features. LIDFS is compared with classic and state-of-the-art methods and a random baseline on 23 high-dimensional datasets, the classic benchmark collection. The evaluation uses classification accuracy and AUC with five downstream classifiers, as well as clustering accuracy and normalized mutual information, over both full and aggressive selection ranges. LIDFS ranks first in the supervised evaluations and first or tied second in the clustering-based evaluations. It belongs to the top statistical group together with methods that require the number of classes, and it runs faster than its closest competitor. Several recent methods, by contrast, fail to outperform the random baseline. A sensitivity analysis shows that small neighborhood sizes yield the best results.
Abstract:High dimensionality degrades both the efficiency and the effectiveness of learning algorithms, and in many applications no class labels are available to guide dimensionality reduction. This paper proposes LIDFS, an unsupervised feature selection method based on Local Intrinsic Dimensionality (LID). The method builds on the formal equivalence between LID and the indiscriminability of distance measures. The data matrix is transposed so that each feature is treated as a point, and the LID of each feature is estimated by maximum likelihood from the distances to its $k$ nearest neighboring features. Features with lower LID exhibit greater local discriminability and receive higher importance scores. A refinement step then retains one representative per feature neighborhood and demotes locally redundant features. The result is a complete ranking from which any number of features can be selected without rerunning the procedure. LIDFS does not use any information about labels or the number of classes, and runs in $\mathcal{O}(d^2 n)$ time for $n$ instances and $d$ features. LIDFS is compared with classic and state-of-the-art methods and a random baseline on 23 high-dimensional datasets, the classic benchmark collection. The evaluation uses classification accuracy and AUC with five downstream classifiers, as well as clustering accuracy and normalized mutual information, over both full and aggressive selection ranges. LIDFS ranks first in the supervised evaluations and first or tied second in the clustering-based evaluations. It belongs to the top statistical group together with methods that require the number of classes, and it runs faster than its closest competitor. Several recent methods, by contrast, fail to outperform the random baseline. A sensitivity analysis shows that small neighborhood sizes yield the best results.
Machine Learning (cs.LG)
arXiv:2610.40736 [cs.LG]
(or arXiv:2610.40736v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.40736 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 365 words
Related coverage
Shape Operator PCA: Curvature-Aware Projections for Geometric Machine Learning
arXiv · Aug 19, 2026
BBOmix: A Tabular Benchmark for Hyperparameter Optimization of Unsupervised Biological Representation Learning
arXiv · Jun 5, 2026
Multimodal Cultural Heritage Architectural Style Classification for Residential Buildings in the UAE Based on CLIP Embeddings and SVM
arXiv · Sep 16, 2026
Hybrid Quantum-Classical NLP Classification with Compact Semantic Representations: An Experimental Analysis of Representation Compression
arXiv · Sep 10, 2026