02901 Advanced Topics in Machine Learning: Tensor Networks for Machine Learning

August 24-28, 2026

Description

The course consists of five days of lectures, exercises, and presentations. The lectures will cover theoretical and practical aspects of tensor networks for machine learning. Technical aspects, programming and application of the covered concepts will be explored in tutorials and exercises. The course (2.5 ECTS points) passed by handing in a small report covering topics in the course. For further course details click here.

Location

DTU Compute building 324 room 060

Confirmed External Speakers

Mahito Sugiyama, associate professor, National Institute of Informatics, Japan.

Evrim Acar, Chief Research Scientist/Research Professor, Simula Metropolitan, Oslo, Norway.

Antonio Vergari , Associate Professor, University of Edinburgh, UK.

Beatriz Quintanilla Casas, Assistant Professor, Department of Food Science, Copenhagen University.

Michael Kastoryano , associate Professor, Department of Computer-science, Copenhagen University.

Sebastian Loeschcke , PhD, Department of Computer-science, Copenhagen University.

Tentative Course Programme (subject to changes)

Monday

Welcome and Course Overview.

Kazu Ghalamkari
Introduction to Tensors and Tensor Networks.

Jesper Løve Hinrich
Applications of Tensor Decompositions to Life Science Data.

Evrim Acar

Tuesday

Tensor Networks as Universal function Approximators, Standard and Bayesian Methods

Niccolo Ciolli
Tensor Networks for Reinforcement Learning

Mads Emil Koefoed Rehof
Tensor Networks in Physics.

Michael Kastoryano
Tensor Decompositions in Chemometrics.

Beatriz Quintanilla Casas

Wednesday

Deep Learning and Tensor Networks.
Petr Taborsky, Morten Mørup
Tensor Networks and Low-Rank Methods for Memory-Efficient Deep Learning

Sebastian Loeschcke

Thursday

Probabilistic circuits and Tensor Networks.
Antonio Vergari
Tensor Networks for Density Estimation Beyond KL Divergence.

Kazu Ghalamkari

Friday

Information-Geometric Modeling for Tensors
Mahito Sugiyama
Project Work and Closing Remarks.

Kazu Ghalamkari, Morten Mørup, Jesper Løve Hinrich

Requirements

General understanding of machine learning, statistical modeling, mathematics and computer science. Programming experience, ideally in Python. For the course you are required to bring your own laptop computer.

Application and Registration

To registrer please send a CV to kazfu@dtu.dk no later than the 27th of June 2026. Confirmations will be sent out by the 30th of June.

After receiving a positive confirmation on the application, registration in DTUs systems must be carried out. For academics (masters and PhD students) there is no registration fee for the course. Students affiliated with DTU can use the course planner to register. PhD students outside of DTU have register via here. For all other participants a course fee will be charged and apart from signing up additional registration must be completed here.