Contrastive Principal Component Analysis
arXiv:1709.06716
Abstract
We present a new technique called contrastive principal component analysis (cPCA) that is designed to discover low-dimensional structure that is unique to a dataset, or enriched in one dataset relative to other data. The technique is a generalization of standard PCA, for the setting where multiple datasets are available -- e.g. a treatment and a control group, or a mixed versus a homogeneous population -- and the goal is to explore patterns that are specific to one of the datasets. We conduct a wide variety of experiments in which cPCA identifies important dataset-specific patterns that are missed by PCA, demonstrating that it is useful for many applications: subgroup discovery, visualizing trends, feature selection, denoising, and data-dependent standardization. We provide geometrical interpretations of cPCA and show that it satisfies desirable theoretical guarantees. We also extend cPCA to nonlinear settings in the form of kernel cPCA. We have released our code as a python package and documentation is on Github.
main body is 10 pages, 9 figures
Cited by in corpus (6)
- Supporting Analysis of Dimensionality Reduction Results with Contrastive Learning
- Nonlinear Dimensionality Reduction for Discriminative Analytics of Multiple Datasets
- Closed-Form Expressions for Global and Local Interpretation of Tsetlin Machines with Applications to Explaining High-Dimensional Data
- Probabilistic Contrastive Principal Component Analysis
- DPCA: Dimensionality Reduction for Discriminative Analytics of Multiple Large-Scale Datasets
- Discriminative community detection for multiplex networks