Jakob Raymaekers
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Jakob Raymaekers has a total of 46 co-authors. Here are the top co-authors by number of shared publications:
- Peter J. Rousseeuw
Section of Statistics and Data Science, KU Leuven, Leuven, Belgium
14 shared publications - Tim Verdonck
Department of Mathematics, University of Antwerp, Antwerp, Belgium
10 shared publications - Ana Helena Tavares
University of Aveiro
3 shared publications - Ines Wilms
Department of Quantitative Economics, Maastricht University
3 shared publications - Paula Brito
Faculty of Economics, University of Porto, Porto, Portugal
3 shared publications - Sarah Leyder
Department of Mathematics, University of Antwerp, Antwerp, Belgium
3 shared publications
Latest Publications
- 23 June 2026
- Published by Springer Nature in Advances in Data Analysis and Classification
- p. 1-35
Fast weighted linear model trees
- 23 June 2026
- Published by Springer Nature in Advances in Data Analysis and Classification
- p. 1-35
Fast weighted linear model trees
- 16 June 2026
- Published by Elsevier in Applied Soft Computing
- Vol. 202
Bayesian optimal experiment design for enhanced mineral weathering with gradient boosted decision trees
- 16 June 2026
- Published by Elsevier in Applied Soft Computing
- Vol. 202
Bayesian optimal experiment design for enhanced mineral weathering with gradient boosted decision trees
- Top Cited PapersOpen Access
- 28 March 2026
- Published by Springer Nature in Advances in Data Analysis and Classification
- Vol. 20 (3) , 1-39
Independent component analysis by robust distance correlation
Independent component analysis (ICA) is a powerful tool that attempts to decompose a multivariate signal or distribution into fully independent sources, not just uncorrelated ones like PCA does. ICA is harder to do, but it has many important applications. Unfortunately, most approaches to ICA are not robust against outliers. Here we propose a robust ICA method called PICARD, which estimates the components by minimizing a robust measure of dependence between multivariate random variables. The dependence measure used is the distance correlation (dCor). In order to make it more robust we first apply a new transformation called the bowl transform, which is bounded, continuous, injective, and maps far outliers to points close to the origin. This preserves the crucial property that a zero dCor implies independence. PICARD estimates the independent sources sequentially, by looking for the component that has the smallest dCor with the remainder. We prove that PICARD is strongly consistent. Its robustness is investigated by a simulation study, in which it generally outperforms its competitors. The method is illustrated on three applications, including the well-known cocktail party problem.
- 28 March 2026
- Published by Springer Nature in Advances in Data Analysis and Classification
- Vol. 20 (3) , 1-39
Top Cited PapersOpen AccessIndependent component analysis by robust distance correlation
Independent component analysis (ICA) is a powerful tool that attempts to decompose a multivariate signal or distribution into fully independent sources, not just uncorrelated ones like PCA does. ICA is harder to do, but it has many important applications. Unfortunately, most approaches to ICA are not robust against outliers. Here we propose a robust ICA method called PICARD, which estimates the components by minimizing a robust measure of dependence between multivariate random variables. The dependence measure used is the distance correlation (dCor). In order to make it more robust we first apply a new transformation called the bowl transform, which is bounded, continuous, injective, and maps far outliers to points close to the origin. This preserves the crucial property that a zero dCor implies independence. PICARD estimates the independent sources sequentially, by looking for the component that has the smallest dCor with the remainder. We prove that PICARD is strongly consistent. Its robustness is investigated by a simulation study, in which it generally outperforms its competitors. The method is illustrated on three applications, including the well-known cocktail party problem.
- Open Access
Robust outlier-adjusted mean-shift estimation of state-space models
State-space models (SSMs) provide a flexible framework for modelling time series data, but their reliance on Gaussian error assumptions makes them highly sensitive to outliers. We propose a robust estimation method, ROAMS, that mitigates the influence of additive outliers by introducing shift parameters at each timepoint in the observation equation of the SSM. These parameters allow the model to attribute non-zero shifts to outliers while leaving clean observations unaffected. ROAMS then enables automatic outlier detection, through the addition of a penalty term on the number of flagged outlying timepoints in the objective function, and simultaneous estimation of model parameters. We apply the method to robustly estimate SSMs on both simulated data and real-world animal location-tracking data, demonstrating its ability to produce more reliable parameter estimates than classical methods and other benchmark methods. In addition to improved robustness, ROAMS offers practical diagnostic tools, including BIC curves for selecting tuning parameters and visualising outlier structure. These features make our approach broadly useful for researchers and practitioners working with contaminated time series data.
Open AccessRobust outlier-adjusted mean-shift estimation of state-space models
State-space models (SSMs) provide a flexible framework for modelling time series data, but their reliance on Gaussian error assumptions makes them highly sensitive to outliers. We propose a robust estimation method, ROAMS, that mitigates the influence of additive outliers by introducing shift parameters at each timepoint in the observation equation of the SSM. These parameters allow the model to attribute non-zero shifts to outliers while leaving clean observations unaffected. ROAMS then enables automatic outlier detection, through the addition of a penalty term on the number of flagged outlying timepoints in the objective function, and simultaneous estimation of model parameters. We apply the method to robustly estimate SSMs on both simulated data and real-world animal location-tracking data, demonstrating its ability to produce more reliable parameter estimates than classical methods and other benchmark methods. In addition to improved robustness, ROAMS offers practical diagnostic tools, including BIC curves for selecting tuning parameters and visualising outlier structure. These features make our approach broadly useful for researchers and practitioners working with contaminated time series data.
- Top Cited PapersOpen Access
- 17 June 2025
- Published by Wiley in Histopathology
- Vol. 87 (5) , 635-646
Practical implementation of AI in a non‐academic, non‐commercial Pathology laboratory: Real world experience and lessons learned
Aims: As pathology departments transition towards digital workflows, the integration of artificial intelligence (AI) is anticipated to become increasingly common. This study aimed to describe the real‐world implementation and impact of AI integration in routine pathological diagnostics, specifically focusing on prostate biopsy evaluations at the Department of Pathology, ZAS Hospitals, Antwerp.Methods and Results: An AI tool for analysing prostate biopsies was integrated into the department's daily workflow by embedding it into existing laboratory information and reporting systems. Following a short adaptation period, the use of AI led to measurable improvements. Most notably, there was a reduction in the number of immunohistochemical tests required, indicating more confident primary diagnoses. Additionally, a significant decrease in turnaround times for biopsy evaluations was observed, highlighting improved efficiency. The implementation process was closely monitored, and practical insights were gathered to guide future AI deployments in pathology.Conclusions: The first‐year experience of integrating AI into daily pathological practice demonstrated tangible benefits in diagnostic efficiency and workflow optimization. However, the process also revealed several challenges related to real‐world deployment, including adaptation by staff and system integration hurdles. The lessons learned provide valuable guidance for other institutions considering similar AI implementations, reinforcing the importance of strategic planning, training and system compatibility in successful adoption.
- 17 June 2025
- Published by Wiley in Histopathology
- Vol. 87 (5) , 635-646
Top Cited PapersOpen AccessPractical implementation of AI in a non‐academic, non‐commercial Pathology laboratory: Real world experience and lessons learned
Aims: As pathology departments transition towards digital workflows, the integration of artificial intelligence (AI) is anticipated to become increasingly common. This study aimed to describe the real‐world implementation and impact of AI integration in routine pathological diagnostics, specifically focusing on prostate biopsy evaluations at the Department of Pathology, ZAS Hospitals, Antwerp.Methods and Results: An AI tool for analysing prostate biopsies was integrated into the department's daily workflow by embedding it into existing laboratory information and reporting systems. Following a short adaptation period, the use of AI led to measurable improvements. Most notably, there was a reduction in the number of immunohistochemical tests required, indicating more confident primary diagnoses. Additionally, a significant decrease in turnaround times for biopsy evaluations was observed, highlighting improved efficiency. The implementation process was closely monitored, and practical insights were gathered to guide future AI deployments in pathology.Conclusions: The first‐year experience of integrating AI into daily pathological practice demonstrated tangible benefits in diagnostic efficiency and workflow optimization. However, the process also revealed several challenges related to real‐world deployment, including adaptation by staff and system integration hurdles. The lessons learned provide valuable guidance for other institutions considering similar AI implementations, reinforcing the importance of strategic planning, training and system compatibility in successful adoption.
- 27 September 2024
- Published by Taylor & Francis in Journal of Computational and Graphical Statistics
- Vol. 34 (2) , 437-447
TSLiNGAM: DirectLiNGAM Under Heavy Tails
- 27 September 2024
- Published by Taylor & Francis in Journal of Computational and Graphical Statistics
- Vol. 34 (2) , 437-447
TSLiNGAM: DirectLiNGAM Under Heavy Tails
- Open Access
- 17 September 2024
- Published by Taylor & Francis in The American Statistician
- Vol. 79 (1) , 122-128
Distance Covariance, Independence, and Pairwise Differences
- 17 September 2024
- Published by Taylor & Francis in The American Statistician
- Vol. 79 (1) , 122-128
Open AccessDistance Covariance, Independence, and Pairwise Differences
- Open Access
- 8 July 2024
- Published by Springer Nature in Machine Learning
- Vol. 113 (9) , 6561-6610
Fast linear model trees by PILOT
Linear model trees are regression trees that incorporate linear models in the leaf nodes. This preserves the intuitive interpretation of decision trees and at the same time enables them to better capture linear relationships, which is hard for standard decision trees. But most existing methods for fitting linear model trees are time consuming and therefore not scalable to large data sets. In addition, they are more prone to overfitting and extrapolation issues than standard regression trees. In this paper we introduce PILOT, a new algorithm for linear model trees that is fast, regularized, stable and interpretable. PILOT trains in a greedy fashion like classic regression trees, but incorporates an L2 boosting approach and a model selection rule for fitting linear models in the nodes. The abbreviation PILOT stands for PIecewise Linear Organic Tree, where ‘organic’ refers to the fact that no pruning is carried out. PILOT has the same low time and space complexity as CART without its pruning. An empirical study indicates that PILOT tends to outperform standard decision trees and other linear model trees on a variety of data sets. Moreover, we prove its consistency in an additive model setting under weak assumptions. When the data is generated by a linear model, the convergence rate is polynomial.
- 8 July 2024
- Published by Springer Nature in Machine Learning
- Vol. 113 (9) , 6561-6610
Open AccessFast linear model trees by PILOT
Linear model trees are regression trees that incorporate linear models in the leaf nodes. This preserves the intuitive interpretation of decision trees and at the same time enables them to better capture linear relationships, which is hard for standard decision trees. But most existing methods for fitting linear model trees are time consuming and therefore not scalable to large data sets. In addition, they are more prone to overfitting and extrapolation issues than standard regression trees. In this paper we introduce PILOT, a new algorithm for linear model trees that is fast, regularized, stable and interpretable. PILOT trains in a greedy fashion like classic regression trees, but incorporates an L2 boosting approach and a model selection rule for fitting linear models in the nodes. The abbreviation PILOT stands for PIecewise Linear Organic Tree, where ‘organic’ refers to the fact that no pruning is carried out. PILOT has the same low time and space complexity as CART without its pruning. An empirical study indicates that PILOT tends to outperform standard decision trees and other linear model trees on a variety of data sets. Moreover, we prove its consistency in an additive model setting under weak assumptions. When the data is generated by a linear model, the convergence rate is polynomial.