Clustering and classification through normalizing flows in feature space

J. P. Agnellit, M. Cadeiras, E. G. Tabak, C. V. Turnert, E. Vanden-Eijnden

Research output: Contribution to journalArticlepeer-review


A unified variational methodology is developed f or classification and clustering problems and is tested in the classification of tumors from gene expression data. It is based on fluid-like flows in feature space that cluster a set of observations by transforming them into likely samples from p isotropic Gaussians, where p is the number of classes sought. The methodology blurs the distinction between training and testing populations through the soft assignment of both to classes. The observations act as Lagrangian markers for the flows, comparatively active or passive depending on the current strength of the assignment to the corresponding class.

Original languageEnglish (US)
Pages (from-to)1784-1802
Number of pages19
JournalMultiscale Modeling and Simulation
Issue number5
StatePublished - 2010


  • Density estimation
  • Expectation maximization
  • Gaussianization
  • Inference
  • Machine learning
  • Maximum likelihood

ASJC Scopus subject areas

  • General Chemistry
  • Modeling and Simulation
  • Ecological Modeling
  • General Physics and Astronomy
  • Computer Science Applications


Dive into the research topics of 'Clustering and classification through normalizing flows in feature space'. Together they form a unique fingerprint.

Cite this