AI Insight
Researchers have developed a new variant of Poisson non-negative matrix factorization (NMF) that uses a shifted-log link function to analyze count data. Unlike standard Poisson NMF methods which assume data components combine additively, this approach allows components to combine in ways ranging from purely additive to more multiplicative, controlled by a single tuning parameter. The team provides efficient algorithms for fitting the model, including an approximation that significantly speeds computation for large, sparse datasets by scaling with non-zero entries rather than total data size.
Why it matters
This method provides greater flexibility in analyzing count data across diverse fields where standard additive assumptions may not hold, potentially improving interpretability of decomposed data patterns. Applications could span genomics, text analysis, recommendation systems, and any domain dealing with count-based observations that exhibit multiplicative rather than additive structure.
Understand the Science
⚠️ Preprint – Noch nicht peer-reviewed
Dieser Artikel wurde noch nicht von unabhängigen Experten begutachtet. Die Ergebnisse sind vorläufig und sollten mit Vorsicht interpretiert werden.
Abstract: Poisson non-negative matrix factorization (NMF) is a widely used method to find interpretable “parts-based” decompositions of count data. While many variants of Poisson NMF exist, existing methods assume that the “parts” in the decomposition combine additively. This assumption may be natural in some settings, but not in others. Here we introduce Poisson NMF with the shifted-log link function to relax this assumption. The shifted-log link function has a single tuning parameter, and as this parameter varies the model changes from assuming that parts combine additively (i.e., standard Poisson NMF) to assuming that parts combine more multiplicatively. We provide an algorithm to fit this model by maximum likelihood, and also an approximation that substantially reduces computation time for large, sparse datasets (computations scale with the number of non-zero entries in the data matrix). We illustrate these new methods on a variety of real datasets. Our examples show how the choice of link function in Poisson NMF can substantively impact the results, and how in some settings the use of a shifted-log link function may improve interpretability compared with the standard, additive link.
Source: A New Family of Poisson Non-negative Matrix Factorization Methods Using the Shifted Log Link