Multi-context principal component analysis
This addresses the challenge of understanding underlying factors in multi-context data across domains like genomics and NLP, offering a novel framework for analysis.
The authors tackled the problem of identifying shared factors across multiple contexts in data, developing Multi-context Principal Component Analysis (MCPCA) to decompose data into factors shared across subsets of contexts, with applications in gene expression revealing axes associated with cancer progression and in language models mapping debate stages.
Principal component analysis (PCA) is a tool to capture factors that explain variation in data. Across domains, data are now collected across multiple contexts (for example, individuals with different diseases, cells of different types, or words across texts). While the factors explaining variation in data are undoubtedly shared across subsets of contexts, no tools currently exist to systematically recover such factors. We develop multi-context principal component analysis (MCPCA), a theoretical and algorithmic framework that decomposes data into factors shared across subsets of contexts. Applied to gene expression, MCPCA reveals axes of variation shared across subsets of cancer types and an axis whose variability in tumor cells, but not mean, is associated with lung cancer progression. Applied to contextualized word embeddings from language models, MCPCA maps stages of a debate on human nature, revealing a discussion between science and fiction over decades. These axes are not found by combining data across contexts or by restricting to individual contexts. MCPCA is a principled generalization of PCA to address the challenge of understanding factors underlying data across contexts.