Supervised Dimensionality Reduction for Big Data

TL;DR

This paper introduces a scalable supervised dimensionality reduction method, LOL, that improves classification accuracy on ultra-high-dimensional biomedical data while maintaining computational efficiency.

Contribution

The paper presents XOX, a novel framework extending PCA with class-conditional moments, and introduces LOL, a simple yet effective supervised reduction technique with theoretical guarantees.

Findings

LOL outperforms existing methods in accuracy on large biomedical datasets

LOL scales efficiently to millions of features within minutes

Theoretical analysis supports the effectiveness of the proposed approach

Abstract

To solve key biomedical problems, experimentalists now routinely measure millions or billions of features (dimensions) per sample, with the hope that data science techniques will be able to build accurate data-driven inferences. Because sample sizes are typically orders of magnitude smaller than the dimensionality of these data, valid inferences require finding a low-dimensional representation that preserves the discriminating information (e.g., whether the individual suffers from a particular disease). There is a lack of interpretable supervised dimensionality reduction methods that scale to millions of dimensions with strong statistical theoretical guarantees.We introduce an approach, XOX, to extending principal components analysis by incorporating class-conditional moment estimates into the low-dimensional projection. The simplest ver-sion, "Linear Optimal Low-rank" projection (LOL), incorporates the class-conditional means. We prove, and substantiate with both synthetic and real data benchmarks, that LOL and its generalizations in the XOX framework lead to improved data representations for subsequent classification, while maintaining computational efficiency and scalability. Using multiple brain imaging datasets consisting of >150 million features, and several genomics datasets with>500,000 features, LOL outperforms other scalable linear dimensionality reduction techniques in terms of accuracy, while only requiring a few minutes on a standard desktop computer.