Gaussian process classification using posterior linearization

Angel F. Garcia-Fernandez*, Filip Tronarp, Simo Särkkä

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

7 Citations (Scopus)


This letter proposes a new algorithm for Gaussian process classification based on posterior linearization (PL). In PL, a Gaussian approximation to the posterior density is obtained iteratively using the best possible linearization of the conditional mean of the labels and accounting for the linearization error. PL has some theoretical advantages over expectation propagation (EP): all calculated covariance matrices are positive definite and there is a local convergence theorem. In experimental data, PL has better performance than EP with the noisy threshold likelihood and the parallel implementation of the algorithms.

Original languageEnglish
Article number8673324
Pages (from-to)735-739
Number of pages5
JournalIEEE Signal Processing Letters
Issue number5
Publication statusPublished - 1 May 2019
MoE publication typeA1 Journal article-refereed


  • Bayesian inference
  • energy efficiency
  • Gaussian process classification
  • posterior linearization
  • simultaneous wireless information and power transfer
  • two-way relaying


Dive into the research topics of 'Gaussian process classification using posterior linearization'. Together they form a unique fingerprint.

Cite this