Niklas Wahlström
Senior Lecturer/Associate Professor at Department of Information Technology; Division of Systems and Control
- Telephone:
- +46 18 471 31 89
- E-mail:
- niklas.wahlstrom@it.uu.se
- Visiting address:
- Hus 10, Regementsvägen 10
- Postal address:
- Box 524
751 20 UPPSALA
- Academic merits:
- Docent in machine learning
- CV:
- Download CV
Short presentation
I am an Associate Professor at the Division of Systems and Control, Department of Information Technology, Uppsala University. My research interests lie in physics-informed machine learning and applications of machine learning in physics.
Refer to my personal home page for more information.
Keywords
- artificial intelligence
- machine learning
- automatic control
- deep learning
- signal processing
- sensor fusion
Biography
Niklas Wahlström is an Associate Professor at the Division of Systems and Control, Department of Information Technology, Uppsala University. His research interests lie in the fields of machine learning, sensor fusion, and statistical signal processing, together with their applications. He is especially interested in physics-informed machine learning and applications of machine learning in physics. He has developed several courses in machine learning, both at MSc level and at PhD level. Niklas received his MSc degree in 2010 and his PhD in automatic control in 2015, both from Linköping University, Sweden. He also did parts of his studies at ETH Zürich (Switzerland) and Imperial College (UK). Since 2016, he has been affiliated with Uppsala University, first as a Postdoctoral researcher, since 2019 as an Assistant Professor, and since 2019 in his present position.

Publications
Recent publications
-
Part of Machine Learning, 2026
- DOI for A differentiable surrogate model for the generation of radio pulses from in-ice neutrino interactions
- Download full text (pdf) of A differentiable surrogate model for the generation of radio pulses from in-ice neutrino interactions
-
Learning dynamics from input-output data with Hamiltonian Gaussian processes.
2026
-
Optimization pipeline for in-ice radio neutrino detectors
Part of Proceedings of Fifth MODE Workshop on Differentiable Programming for Experiment Design PoS(MODE2025), 2025
-
Physics-informed neural networks with unknown measurement noise
Part of Proceedings of Machine Learning Research, p. 235-247, 2024
-
Probabilistic Matching of Real and Generated Data Statistics in Generative Adversarial Networks
Part of Transactions on Machine Learning Research, 2024
All publications
Articles in journal
-
Part of Machine Learning, 2026
- DOI for A differentiable surrogate model for the generation of radio pulses from in-ice neutrino interactions
- Download full text (pdf) of A differentiable surrogate model for the generation of radio pulses from in-ice neutrino interactions
-
Probabilistic Matching of Real and Generated Data Statistics in Generative Adversarial Networks
Part of Transactions on Machine Learning Research, 2024
-
Invertible Kernel PCA With Random Fourier Features
Part of IEEE Signal Processing Letters, p. 563-567, 2023
-
Incorporating Sum Constraints into Multitask Gaussian Processes
Part of Transactions on Machine Learning Research, p. 1-28, 2022
-
Probabilistic approach to limited-data computed tomography reconstruction
Part of Inverse Problems, 2019
-
Probabilistic modelling and reconstruction of strain
Part of Nuclear Instruments and Methods in Physics Research Section B, p. 141-155, 2018
-
Modeling and Interpolation of the Ambient Magnetic Field by Gaussian Processes
Part of IEEE Transactions on robotics, p. 1112-1127, 2018
-
A Platform for Teaching Sensor Fusion Using a Smartphone
Part of International journal of engineering education, p. 781-789, 2017
Books
-
Machine learning: a first course for engineers and scientists
Cambridge University Press, 2022
Conference papers
-
Learning dynamics from input-output data with Hamiltonian Gaussian processes.
2026
-
Optimization pipeline for in-ice radio neutrino detectors
Part of Proceedings of Fifth MODE Workshop on Differentiable Programming for Experiment Design PoS(MODE2025), 2025
-
Physics-informed neural networks with unknown measurement noise
Part of Proceedings of Machine Learning Research, p. 235-247, 2024
-
Learning deep autoregressive models for hierarchical data
Part of IFAC PapersOnLine, p. 529-534, 2021
- DOI for Learning deep autoregressive models for hierarchical data
- Download full text (pdf) of Learning deep autoregressive models for hierarchical data
-
First Steps Towards Self-Supervised Pretraining of the 12-Lead ECG
Part of 2021 Computing In Cardiology (CINC), 2021
-
Deep State Space Models for Nonlinear System Identification
Part of IFAC PapersOnLine, p. 481-486, 2021
- DOI for Deep State Space Models for Nonlinear System Identification
- Download full text (pdf) of Deep State Space Models for Nonlinear System Identification
-
Deep convolutional networks in system identification
Part of Proc. 58th IEEE Conference on Decision and Control, p. 3670-3676, 2019
-
Data-driven impulse response regularization via deep learning
p. 1-6, 2018
-
Linearly constrained Gaussian processes
Part of Proc. 31st Conference on Neural Information Processing Systems, p. 1215-1224, 2017
-
Rao-Blackwellised Particle Filter for Star-ConvexExtended Target Tracking Models
Part of 2016 19th International Conference on Information Fusion, p. 1193-1199, 2016