PhD position in probabilistic machine learning and statistics

3-years PhD position in probabilistic machine learning and statistics

We invite applications for a three-year PhD Research Fellowship in probabilistic machine learning and statistics at theOslo Centre for Biostatistics and Epidemiology (OCBE) , Department of Biostatistics, Institute of Basic Medical Sciences (IMB), University of Oslo (UiO), Norway.

The preferred starting date is as soon as possible and will be agreed upon with the successful candidate.

No one can be appointed for more than one PhD Research Fellowship period at the University of Oslo.

Place of work is the Department of Biostatistics (OCBE), Domus Medica, Gaustad UiO campus, Oslo.

The project focuses on developing probabilistic latent-variable methods for large and structured biological data, with applications in genomics, spatial transcriptomics, and fluorescence imaging.

High-dimensional and structured biological data are increasingly common in modern biomedical research but remain challenging to analyse because of their scale, heterogeneity, and complex spatial and functional dependencies. Existing methods often rely on restrictive assumptions or application-specific computational workflows. The PhD project aims to address these limitations by developing a unified, scalable, and interpretable framework for probabilistic unsupervised learning for structured biological data.

The successful candidate will:

• Develop probabilistic factor models and scalable inference algorithms for structured biological (multi-view) high-dimensional data.
• Develop modular methods that incorporate domain-specific information into latent-variable models.
• Investigate methodological questions related to computation, identifiability, uncertainty quantification, and interpretation.
• Account for structures arising from spatial relationships, physical constraints, high-dimensional imaging, and clinical covariates.
• Apply the resulting methods to spatial transcriptomics and fluorescence imaging data to improve our understanding of complex biological systems.

The project is particularly suited to a candidate interested in probabilistic modelling, latent-variable methods, and structured unsupervised learning.

Research Environment & Collaboration

The successful candidate will work at the interface of probabilistic machine learning, computational statistics, and biostatistics, developing new methodology, inference algorithms, and scalable implementations. By contributing to a new class of structured factor model, the candidate will work on foundational methodological questions motivated by complex biomedical data.

• Global Impact:You will join the
  Fun Gen-AD consortium
  , the world’s largest research initiative studying the genetic underpinnings of Alzheimer’s disease.
• International Mobility:there are possibilities for arranging a3 to 6-month research stay
  at Columbia University in New York (USA).
• Publication:Candidates are encouraged to publish in top-tier venues across machine learning (e.g., NeurIPS, ICML), statistics, and computational biology.
• Dual Affiliation:The position will be based at and affiliated with the University of Oslo (Norway) and will also be affiliated with
  Columbia University (USA).

The research group on statistical models for high-dimensional and functional data is part of the larger and active research environment on “High-dimensional statistics” E has expanded considerably during the last decade, becoming one of Europe's most active biostatistics groups with currently over 70 researchers. OCBE is internationally recognized, with interests spanning a broad range of research areas - including methods for high-dimensional data and data integration, especially in molecular medicine;

mathematical modelling of cancer; probabilistic modelling and Bayesian…