My current research is mainly focused on topological and geometric machine learning and its application in medicine and energy. Previously I have worked on topics ranging from epidemiology, toxicology and geopyhsics to statistics and topological data analysis.
One important method in topological data analysis is persistent homology. The method consists of several steps: First, the data is translated into a (bi-)filtered simplicial complex. Starting from low filtration values, this filtered simplicial complex is then assembled, while keeping track of the components, holes and voids in the structure. In this way, we keep track of the features and the range of filtration values for which they exist, allowing for multi-scale analysis. Features that persist over large ranges of filtration values are called persistent features and thought to be important. For unsupervised machine learning, we can then visualize the persistence diagrams and localize some homology features. For supervised machine learning, persistence diagrams need to be transformed into a vector representation, before using standard supervised learning algorithms. My research spans from encoding of data into (bi-)filtered simplicial complexes (topological representation), approximations of filtered simplicial complexes (sparsification), localizing topological features, validation measures of unsupervised persistent homology, data benchmarks, efficient vector representations of persistent homology for supervised machine learning and applications of topological methods in the domains of biomedicine, geophysics and energy.
Another important tool from topology are Reeb graphs, which can be used to characterize manifolds. Since manifold learning is an important but challenging part of machine learning, Reeb graph learning can contribute by capturing some of the important properties of the manifold. My research spans the following different aspects of Reeb graph learning: mathematical foundations, (probabilistic) learning algorithms, performance measures, metrics, visualization, downstream tasks such as dimensionality reduction or generative modelling and applications to different domains.
Geometric deep learning provides a principled framework for designing neural networks that respect the symmetries of the underlying data domain. When data is defined on a structured space such as the sphere in the case of global atmospheric fields, incorporating the corresponding symmetry group as an inductive bias leads to more data-efficient and generalizable models. My research in this area spans the development of equivariant methods for previously unexplored domains and symmetries.
In the context of machine learning applications I am interested in understanding how spatial structures and their change over time influence biology, energy systems, and epidemics. Some examples of previous research includes studying the spread of HIV, the effects of various toxins on fish development, and rapid treatment responses in leukemia.
I am currently involved in the following funded research projects:
I have previously been involved in the following funded research projects:
Below you find a list of preprints.
[1] J. S. Schmidt, M. Carrasco, E. Röell, G. Wolf, N. Blaser, and B. Rieck, “No triangulation without representation: Generalization in topological deep learning.” 2026, url: https://arxiv.org/abs/2605.06467.
[2] M. Blørstad, H. J. Mostein, N. Blaser, and P. Parviainen, “Evaluating prediction uncertainty estimates from BatchEnsemble.” 2026, url: https://arxiv.org/abs/2601.21581.
[3] H. B. Bjerkevik, N. Blaser, and L. M. Salbu, “Reeb graph of sample thickenings.” 2025, url: https://arxiv.org/abs/2512.08159.
[4] S. Gavasso et al., “High-dimensional immune profiling following autologous hematopoietic stem cell transplantation in relapsing-remitting multiple sclerosis,” 2025, doi: 10.1101/2025.07.01.662494.
[5] M. H. Hannisdal et al., “Spatial-temporal recurrence patterns of grade 4 glioma using deep learning integrated mpMRI and molecular pathology: A multi-centre observational study,” 2025, doi: http://dx.doi.org/10.2139/ssrn.5177606.
[6] F. Ballerin, N. Blaser, and E. Grong, “SO(3)-equivariant neural networks for learning from scalar and vector fields on spheres.” 2025, url: https://arxiv.org/abs/2503.09456.
Below you find a list of peer-reviewed publications.
[1] N. Blaser, M. Brun, O. H. Gardaa, and L. M. Salbu, “Monoidal Rips: Stable multiparameter filtrations of directed networks,” Journal of Applied and Computational Topology, vol. 10, no. 2, 2026, doi: 10.1007/s41468-026-00237-z.
[2] A. Grefsrud, T. Buanes, and N. Blaser, “Calibrated and uncertain? Evaluating uncertainty estimates in binary classification models,” Machine Learning: Science and Technology, vol. 7, no. 2, p. 025016, Mar. 2026, doi: 10.1088/2632-2153/ae45ed.
[3] O. H. Gardaa and N. Blaser, “RotaTouille: Rotation equivariant deep learning for contours,” 2025, url: https://openreview.net/forum?id=VbCLyz3uW7.
[4] N. Blaser, M. Brun, O. H. Gardaa, and L. M. Salbu, “Core bifiltration,” Journal of Applied and Computational Topology, vol. 9, no. 4, 2025, doi: 10.1007/s41468-025-00226-8.
[5] N. Blaser and E. R. Vågset, “Homology localization through the looking-glass of parameterized complexity theory,” Journal of Applied and Computational Topology, vol. 9, no. 2, 2025, doi: 10.1007/s41468-025-00212-0.
[6] B. Bergougnoux, N. Blaser, M. Fellows, P. Golovach, F. Rosamond, and E. Sam, “On the parameterized complexity of lineal topologies (depth-first spanning trees) with many or few leaves,” Journal of Computer and System Sciences, vol. 154, p. 103680, 2025, doi: https://doi.org/10.1016/j.jcss.2025.103680.
[7] M. Blørstad, B. Å. S. Lunde, and N. Blaser, “Stable update of regression trees,” in Proceedings of the 3rd conference on lifelong learning agents, 2025, vol. 274, pp. 641–651, url: https://proceedings.mlr.press/v274/blorstad25a.html.
[8] N. Blaser, M. Brun, L. M. Salbu, and E. R. Vågset, “The parameterized complexity of finding minimum bounded chains,” Computational Geometry, vol. 122, p. 102102, 2024, doi: https://doi.org/10.1016/j.comgeo.2024.102102.
[9] E. Sam, B. Bergougnoux, P. A. Golovach, and N. Blaser, “Kernelization for finding lineal topologies (depth-first spanning trees) with many or few leaves,” in Fundamentals of computation theory, 2023, pp. 392–405, doi: 10.1007/978-3-031-43587-4_28.
[10] M. Eide et al., “Integrative omics-analysis of lipid metabolism regulation by peroxisome proliferator-activated receptor a and b agonists in male atlantic cod (gadus morhua),” Frontiers in Physiology, vol. 14, p. 437, 2023, doi: 10.3389/fphys.2023.1129089.
[11] B. Tislevoll et al., “Early response evaluation by single cell signaling profiling in acute myeloid leukemia,” Nature Communications, vol. 14, no. 1, p. 115, Jan. 2023, doi: 10.1038/s41467-022-35624-4.
[12] S. Mayala et al., “GUBS: Graph-based unsupervised brain segmentation in MRI images,” Journal of Imaging, vol. 8, no. 10, 2022, doi: 10.3390/jimaging8100262.
[13] N. Blaser and M. Brun, “Relative Persistent Homology,” Discrete & Computational Geometry, 2022, doi: 10.1007/s00454-022-00421-9.
[14] M. Black, N. Blaser, A. Nayyeri, and E. R. Vågset, “ETH-Tight Algorithms for Finding Surfaces in Simplicial Complexes of Bounded Treewidth,” in 38th international symposium on computational geometry (SoCG 2022), 2022, vol. 224, pp. 17:1–17:16, doi: 10.4230/LIPIcs.SoCG.2022.17.
[15] G. H. Bringeland, N. Blaser, K.-M. Myhr, C. A. Vedeler, and S. Gavasso, “Wearing-off symptoms during standard and extended natalizumab dosing intervals: Experiences from the COVID-19 pandemic,” Journal of the Neurological Sciences, p. 117622, 2021, doi: 10.1016/j.jns.2021.117622.
[16] N. Galmiche, H. Hauser, T. Spengler, C. Spensberger, M. Brun, and N. Blaser, “Revealing Multimodality in Ensemble Weather Prediction,” in Machine learning methods in visualisation for big data, 2021, doi: 10.2312/mlvis.20211073.
[17] K. Gundersen, A. Oleynik, N. Blaser, and G. Alendal, “Semi-conditional variational auto-encoder for flow reconstruction and uncertainty quantification from limited observations,” Physics of Fluids, vol. 33, no. 1, p. 017119, 2021, doi: 10.1063/5.0025779.
[18] N. Blaser and M. Aupetit, “Research directions to validate topological models of multi-dimensional data,” 2020, url: https://openreview.net/forum?id=3iVbgfPMZtU.
[19] K. Gundersen, G. Alendal, A. Oleynik, and N. Blaser, “Binary time series classification with bayesian convolutional neural networks when monitoring for marine gas discharges,” Algorithms, vol. 13, p. 145, June 2020, doi: 10.3390/a13060145.
[20] N. Blaser and M. Brun, “Relative Persistent Homology,” in 36th international symposium on computational geometry (SoCG 2020), 2020, vol. 164, pp. 18:1–18:10, doi: 10.4230/LIPIcs.SoCG.2020.18.
[21] A. Oleynik, M. I. García-Ibáñez, N. Blaser, A. Omar, and G. Alendal, “Optimal sensors placement for detecting CO2 discharges from unknown locations on the seafloor,” International Journal of Greenhouse Gas Control, vol. 95, p. 102951, 2020, doi: 10.1016/j.ijggc.2019.102951.
[22] G. H. Bringeland, N. Blaser, K.-M. Myhr, C. A. Vedeler, and S. Gavasso, “Wearing-off at the end of natalizumab dosing intervals is associated with low receptor occupancy,” Neurology - Neuroimmunology Neuroinflammation, vol. 7, no. 3, 2020, doi: 10.1212/NXI.0000000000000678.
[23] N. Blaser and M. Brun, “Sparse nerves in practice,” in Machine learning and knowledge extraction, 2019, pp. 272–284, doi: 10.1007/978-3-030-29726-8_17.
[24] L. Bader et al., “Candidate markers for stratification and classification in rheumatoid arthritis,” Frontiers in Immunology, vol. 10, p. 1488, 2019, doi: 10.3389/fimmu.2019.01488.
[25] M. Brun and N. Blaser, “Sparse dowker nerves,” Journal of Applied and Computational Topology, pp. 1–28, 2019, doi: 10.1007/s41468-019-00028-9.
[26] K. Dale et al., “Contaminant accumulation and biological responses in atlantic cod (gadus morhua) caged at a capped waste disposal site in kollevåg, western norway,” Marine Environmental Research, 2019, doi: 10.1016/j.marenvres.2019.02.003.
[27] G. H. Bringeland et al., “Optimization of receptor occupancy assays in mass cytometry: Standardization across channels with QSC beads,” Cytometry Part A, 2019, doi: 10.1002/cyto.a.23723.
[28] K. E. M. Ahmed et al., “An LC-MS/MS approach to reveal the effects of POPs and endocrine disruptors on the steroidogenesis of the human H295R adrenocortical cell line,” Chemosphere, 2019, doi: 10.1016/j.chemosphere.2018.11.057.
[29] H. G. Frøysa, S. Fallahi, and N. Blaser, “Evaluating model reduction under parameter uncertainty,” BMC Systems Biology, vol. 12, no. 1, p. 79, 2018, doi: 10.1186/s12918-018-0602-x.
[30] N. Blaser and M. Brun, “Divisive cover,” Mathematics in Computer Science, vol. 13, no. 1, pp. 21–29, 2019, doi: 10.1007/s11786-018-0352-6.
[31] F. Yadetie et al., “RNA-seq analysis of transcriptome responses in atlantic cod (gadus morhua) precision-cut liver slices exposed to benzo[a]pyrene and 17–ethynylestradiol,” Aquatic Toxicology, vol. 201, pp. 174–186, 2018, doi: 10.1016/j.aquatox.2018.06.003.
[32] J. Estill et al., “The effect of monitoring viral load and tracing patients lost to follow-up on the course of the HIV epidemic in malawi: A mathematical model,” Open Forum Infectious Diseases, vol. 5, no. 5, p. ofy092, 2018, doi: 10.1093/ofid/ofy092.
[33] N. Blaser et al., “Impact of screening and ART on anal cancer incidence in HIV-positive men who have sex with men: Mathematical modeling study,” AIDS, 2017, doi: 10.1097/QAD.0000000000001546.
[34] J. Estill et al., “Estimating the need of second-line antiretroviral therapy in sub-Saharan Africa up to 2030: A mathematical model,” The Lancet HIV, vol. 3, no. 3, pp. e132–e139, 2016, doi: 10.1016/S2352-3018(16)00016-3.
[35] N. Blaser et al., “Tuberculosis in cape town: An age-structured transmission model,” Epidemics, vol. 14, pp. 54–61, 2016, doi: 10.1016/j.epidem.2015.10.001.
[36] O. Keiser et al., “Growth in Virologically Suppressed HIV Positive Children on Antiretroviral Therapy: Individual and Population-Level References,” The Pediatric Infectious Disease Journal, 2015, doi: 10.1097/INF.0000000000000801.
[37] J. Estill, L. Salazar-Vizcaya, N. Blaser, M. Egger, and O. Keiser, “The Cost-Effectiveness of Monitoring Strategies for Antiretroviral Therapy of HIV Infected Patients in Resource-Limited Settings: Software Tool,” PLoS ONE, vol. 10, no. 3, p. e0119299, 2015, doi: 10.1371/journal.pone.0119299.
[38] N. Blaser et al., “gems: An R package for simulating from disease progression models,” Journal of Statistical Software, vol. 64, no. 10, pp. 1–22, 2015, doi: 10.18637/jss.v064.i10.
[39] L. Salazar-Vizcaya et al., “Viral load versus CD4+ monitoring and 5-year outcomes of antiretroviral therapy in HIV-positive children in Southern Africa: A cohort-based modelling study,” AIDS, vol. 28, no. 16, pp. 2451–2460, 2014, doi: 10.1097/QAD.0000000000000446.
[40] M. Petersen, J. Schwab, S. Gruber, N. Blaser, M. Schomaker, and M. van der Laan, “Targeted maximum likelihood estimation for dynamic and static longitudinal marginal structural working models,” Journal of Causal Inference, vol. 2, no. 2, pp. 147–185, 2014, doi: 10.1515/jci-2013-0007.
[41] N. Blaser et al., “Impact of viral load and the duration of primary infection on HIV transmission: Systematic review and meta-analysis,” AIDS, vol. 28, no. 7, pp. 1021–1029, 2014, doi: 10.1097/QAD.0000000000000135.
[42] J. Estill et al., “Tracing of patients lost to follow-up and HIV transmission: Mathematical modeling study based on 2 large ART programs in Malawi,” Journal of acquired immune deficiency syndromes, vol. 65, no. 5, pp. e179–186, 2014, doi: 10.1097/QAI.0000000000000075.
[43] D. Keebler et al., “Cost-effectiveness of different strategies to monitor adults on antiretroviral treatment: A combined analysis of three mathematical models,” The Lancet Global Health, vol. 2, no. 1, pp. e35–e43, 2014, doi: 10.1016/S2214-109X(13)70048-2.
[44] J. Estill et al., “Cost-effectiveness of point-of-care viral load monitoring of antiretroviral therapy in resource-limited settings: Mathematical modelling study,” AIDS, vol. 27, no. 9, pp. 1483–1492, 2013, doi: 10.1097/QAD.0b013e328360a4e5.
[45] C. Wettstein et al., “Missed opportunities to prevent mother-to-child-transmission: Systematic review and meta-analysis,” AIDS, vol. 26, no. 18, pp. 2361–2373, 2012, doi: 10.1097/QAD.0b013e328359ab0c.