PhD in Generative AI and Statistics: Coalescent-inspired Diffusion Models for Discrete Data University of Oslo in Norway

The University of Oslo, Norway’s oldest and highest ranked educational and research institution, invites applications for a PhD position in Generative AI and Statistics, focusing on coalescent-inspired diffusion models for discrete data. The successful candidate will join Integreat – the Norwegian Centre for Knowledge-driven Machine Learning, a Centre of Excellence funded by the Research Council of Norway. Integreat is renowned for its interdisciplinary research, drawing expertise from mathematics, informatics, philosophy, biostatistics, and computer science, and offers a vibrant, international academic environment.

This doctoral research fellowship is based at the Institute of Basic Medical Sciences, with the place of work at Integreat in Oslo. The project aims to develop a novel, mathematically principled generative modeling framework for discrete sequence data by unifying diffusion-based generative modeling with coalescent theory from population genetics. The central idea is to replace heuristic discrete denoising schemes with coalescent-inspired stochastic processes, leveraging the duality between forward allele-frequency diffusion and backward genealogical merging processes. The research will position the candidate at the forefront of modern generative AI, with opportunities to publish in top-tier venues in both machine learning and statistics.

The PhD fellowship is for three years, with a possible fourth year involving 25% career-promoting work such as teaching, supervision, or research assistance, depending on qualifications and departmental needs. The position offers a unique and ambitious research environment, access to a strong network of national and international collaborators, structured career development programs, mentoring, research mobility funds, and a family-friendly, flexible working environment. Salary ranges from NOK 550,800 to 595,000 per year, with pension and welfare benefits, and full access to public health services.

Applicants must have a Master’s degree (120 ECTS) or equivalent in machine learning, statistics, mathematics, computer science, physics, or a closely related quantitative discipline, with a minimum grade B (ECTS grading scale) or equivalent. The degree must include a thesis of at least 30 ECTS. Required skills include competence in probability, linear algebra, statistical modelling, proficiency in programming (Python, PyTorch/JAX, or similar), and strong interest in method development. Fluent oral and written English is essential. Candidates without a master’s degree must complete the final exam by 01.09.2026. Desired qualifications include Bayesian statistics, empirical Bayes methods, advanced probabilistic modelling, familiarity with stochastic processes, prior exposure to Transformer architectures or large-scale sequence modelling, and experience with diffusion models.

Personal skills such as analytical ability, motivation for theory-driven research, independence, collaboration, persistence, creativity, and ambition for an academic or research-oriented career are valued. Integreat is committed to equity, diversity, inclusion, and belonging, with structured evaluations, targeted mentoring, and flexible work arrangements. Applications from women, gender minorities, and candidates with diverse backgrounds are particularly welcomed.

To apply, submit your application via Jobbnorge, including a cover letter, CV, educational certificates, documentation of English proficiency (if applicable), publication list, and 2-3 references. Name documents as 'Document type - Surname - First name'. Foreign applicants should attach an explanation of their university's grading system. All documentation must be in English or a Scandinavian language. The deadline for applications is 11th May 2026.

For further information, contact William Denault (Researcher) or Valeria Vitelli (Professor) at the provided emails. The best qualified candidates will be invited for interviews. The University of Oslo is committed to gender equality and diversity, with dedicated initiatives and networks for women in science.