A locally adaptive normal distribution

Georgios Arvanitidis, Lars Kai Hansen, Søren Hauberg

Research output: Contribution to journalConference articleResearchpeer-review

244 Downloads (Pure)

Abstract

The multivariate normal density is a monotonic function of the distance to the mean, and its ellipsoidal shape is due to the underlying Euclidean metric. We suggest to replace this metric with a locally adaptive, smoothly changing (Riemannian) metric that favors regions of high local density. The resulting locally adaptive normal distribution (LAND) is a generalization of the normal distribution to the "manifold" setting, where data is assumed to lie near a potentially low-dimensional manifold embedded in RD. The LAND is parametric, depending only on a mean and a covariance, and is the maximum entropy distribution under the given metric. The underlying metric is, however, non-parametric. We develop a maximum likelihood algorithm to infer the distribution parameters that relies on a combination of gradient descent and Monte Carlo integration. We further extend the LAND to mixture models, and provide the corresponding EM algorithm. We demonstrate the efficiency of the LAND to fit non-trivial probability distributions over both synthetic data, and EEG measurements of human sleep.
Original languageEnglish
JournalAdvances in Neural Information Processing Systems
Pages (from-to)4258-4266
Number of pages9
ISSN1049-5258
Publication statusPublished - 2016
Event30th Annual Conference on Neural Information Processing Systems - Centre Convencions Internacional Barcelona, Barcelona, Spain
Duration: 5 Dec 201610 Dec 2016

Conference

Conference30th Annual Conference on Neural Information Processing Systems
LocationCentre Convencions Internacional Barcelona
CountrySpain
CityBarcelona
Period05/12/201610/12/2016

Keywords

  • Computer Networks and Communications
  • Information Systems
  • Signal Processing
  • Maximum likelihood
  • Normal distribution
  • Distribution parameters
  • Euclidean metrics
  • Low-dimensional manifolds
  • Maximum entropy distribution
  • Maximum likelihood algorithm
  • Monotonic functions
  • Monte Carlo integration
  • Multivariate normal
  • Probability distributions

Fingerprint Dive into the research topics of 'A locally adaptive normal distribution'. Together they form a unique fingerprint.

Cite this