References & Citations
Statistics > Machine Learning
Title: Linear Optimal Low Rank Projection Provably Outperforms Principal Components Analysis in High-Dimensional Multi-class Data
(Submitted on 5 Sep 2017 (this version), latest version 23 Jan 2021 (v9))
Abstract: Classification of individual samples into one or more categories is critical to modern scientific inquiry. Most modern datasets, such as those used in genetic analysis or imaging, include numerous features, such as genes or pixels. Principal Components Analysis (PCA) is now generally used to find low-dimensional representations of such features for further analysis. However, PCA ignores class label information, thereby discarding data that could substantially improve downstream classification performance. We here describe an approach called "Linear Optimal Low-rank"' projection (LOL), which extends PCA by incorporating the class labels. Using theory and synthetic data, we show that LOL leads to a better representation of the data for subsequent classification than PCA while adding negligible computational cost. Experimentally we demonstrate that LOL substantially outperforms PCA in differentiating cancer patients from healthy controls using genetic data and in differentiating gender from magnetic resonance imaging data incorporating >500 million features and 400 gigabytes of data. LOL allows the solution of previous intractable problems yet requires only a few minutes to run on a single desktop computer.
Submission history
From: Eric Bridgeford [view email][v1] Tue, 5 Sep 2017 04:19:38 GMT (164kb,D)
[v2] Wed, 6 Sep 2017 05:32:45 GMT (176kb,D)
[v3] Tue, 12 Sep 2017 02:32:39 GMT (176kb,D)
[v4] Wed, 13 Sep 2017 06:07:46 GMT (176kb,D)
[v5] Tue, 27 Feb 2018 05:28:42 GMT (922kb,D)
[v6] Wed, 21 Nov 2018 10:17:19 GMT (448kb,D)
[v7] Tue, 7 Jul 2020 09:44:03 GMT (490kb,D)
[v8] Mon, 13 Jul 2020 16:01:37 GMT (494kb,D)
[v9] Sat, 23 Jan 2021 07:37:55 GMT (600kb,D)
Link back to: arXiv, form interface, contact.