Probabilistic Machine Learning

We develop new methods for machine learning, computational inference, and probabilistic modeling. We focus on models for learning from multiple data sources, including multi-view learning, multi-task learning, and multi-way learning, and methods combining mechanistic models and probabilistic inference. Our primary application areas are computational systems biology and medicine, bioinformatics, proactive information retrieval and multimodal interfaces, as well as brain signal analysis and neuroinformatics.

Our main lines of work include Bayesian methodology for modeling dependencies between multiple data sets with co-occurring samples, model-driven methods for retrieving and visualizing data sets, and Gaussian process differential equation models and inference methods for short genomic time series data.




The principal investigators of the research group are Samuel Kaski (Professor, Group Leader), Antti Honkela (Docent, Academy Research Fellow) and Jaakko Peltonen (Docent, Academy Research Fellow). The group operates at

  • Aalto University, Department of Information and Computer Science - PIs Samuel Kaski and Jaakko Peltonen (see the Aalto group website for details),
  • University of Helsinki, Department of Computer Science - PIs Samuel Kaski and Antti Honkela.

We are members of


Principal investigators



Doctoral students


Last updated on 18 Dec 2015 by Tommi Mononen - Page created on 13 Jan 2007 by Webmaster