Marco Loog
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
Marco Loog has a total of 104 co-authors. Here are the top co-authors by number of shared publications:
- Jesse H. Krijthe
Delft University of Technology, Delft, Netherlands
13 shared publications - Tom Viering
Delft University of Technology, Delft, Netherlands
11 shared publications - Alexander Mey
Delft University of Technology
7 shared publications - David M. J. Tax
Delft University of Technology
4 shared publications - David M. J. Tax
Delft University of Technology, Delft, The Netherlands
4 shared publications - Lodewyk F. A. Wessels
The Netherlands Cancer Institute, Amsterdam, The Netherlands
4 shared publications
Latest Publications
- Open Access
- 22 January 2026
- Published by SAGE Publications in Journal of Alzheimer's Disease Reports
- Vol. 10
Applications and implicit assumptions in dementia risk scores: A scoping review of the LIBRA score
Dementia risk scores are commonly used tools to estimate the risk of developing Alzheimer's disease and dementia. We lack an overview of what risk scores are used for, what is claimed they ought to be used for, and whether they are suitable for these applications. To address this, we use the ‘Lifestyle for Brain Health’ (LIBRA) score as a representative example risk score and conduct a literature review to study its applications. The goals of this study are (1) to create an overview of how the LIBRA score has been utilized in scientific articles, (2) to record other applications that these same articles mention, and (3) to critically assess whether LIBRA is suitable for these applications. Of the 66 articles included in our review, 36 involved analyzing associations of LIBRA with dementia, cognition, or other outcomes. We also identified several other applications, with 32 articles mentioning LIBRA as an estimate of ‘dementia prevention potential’, 6 articles used LIBRA as a surrogate outcome for their trial or intervention, and 7 articles mentioned that it could help support clinician decisions in practice. Although there is a clear need for tools that can be used for these applications, the amount of evidence supporting the suitability of dementia risk scores for many of these applications is limited. We recommend that researchers transparently report the purposes of these dementia risk scores, which may include causal tasks, and that research is done to evaluate whether it is valid to use these scores in this way.
- 22 January 2026
- Published by SAGE Publications in Journal of Alzheimer's Disease Reports
- Vol. 10
Open AccessApplications and implicit assumptions in dementia risk scores: A scoping review of the LIBRA score
Dementia risk scores are commonly used tools to estimate the risk of developing Alzheimer's disease and dementia. We lack an overview of what risk scores are used for, what is claimed they ought to be used for, and whether they are suitable for these applications. To address this, we use the ‘Lifestyle for Brain Health’ (LIBRA) score as a representative example risk score and conduct a literature review to study its applications. The goals of this study are (1) to create an overview of how the LIBRA score has been utilized in scientific articles, (2) to record other applications that these same articles mention, and (3) to critically assess whether LIBRA is suitable for these applications. Of the 66 articles included in our review, 36 involved analyzing associations of LIBRA with dementia, cognition, or other outcomes. We also identified several other applications, with 32 articles mentioning LIBRA as an estimate of ‘dementia prevention potential’, 6 articles used LIBRA as a surrogate outcome for their trial or intervention, and 7 articles mentioned that it could help support clinician decisions in practice. Although there is a clear need for tools that can be used for these applications, the amount of evidence supporting the suitability of dementia risk scores for many of these applications is limited. We recommend that researchers transparently report the purposes of these dementia risk scores, which may include causal tasks, and that research is done to evaluate whether it is valid to use these scores in this way.
- Open Access
- 3 January 2026
- Published by Elsevier in Pattern Recognition Letters
- Vol. 201, 29-36
Generalization performance distributions along learning curves
- 3 January 2026
- Published by Elsevier in Pattern Recognition Letters
- Vol. 201, 29-36
Open AccessGeneralization performance distributions along learning curves
- 30 September 2025
- Published by Springer Nature
- p. 362-379
The Vanishing Empirical Variance in Randomly Initialized Deep ReLU Networks
- 30 September 2025
- Published by Springer Nature
- p. 362-379
The Vanishing Empirical Variance in Randomly Initialized Deep ReLU Networks
- Open Access
- 9 July 2025
- Published by American Chemical Society (ACS) in ChemRxiv
Fractionation of Lignin with Aqueous Organic Solvents: A Step Closer to Sustainable Wood Biorefinery
Recent rapid developments in forest biomass valorisation have highlighted lignin as a key value driver for the economic sustainability of modern biorefining. Sugar-based biorefinery hydrolysis lignin (HL) from wood biomass has an immense potential as an alternative for fossil-based aromatic feedstock. However, impurities, heterogeneous structures and limited solubility in common solvents are the main challenges associated with HL functionalisation. Aqueous organic solvent fractionation represents a promising approach to improve on HL properties and subsequently increase its valorisation options. In this study, for the first time, biorefinery HL derived from birch was purified and fractionated with water-EtOH and -THF mixtures across the whole co-solvent concentration range at ambient temperature. The highest lignin solubility was achieved with 70 wt% THF mixture, which dissolved a remarkable 81 wt% of lignin available in HL. Moreover, fractionated lignin purity was further increased to 95 wt% when employing an extra water-washing step. In general, fractions soluble in THF mixtures had high dispersity, syringyl-to-guaiacyl (S/G) subunit ratio and aliphatic OH-groups, whereas EtOH fractions had narrower molecular weight distributions, lower S/G subunit ratio and higher concentration of phenolic OH-groups. A brief description of the solvent effects at the molecular level was given using molecular dynamics simulations which revealed specific solvation between the dissolved lignin and solvent mixture. Protocol presented in this paper enables to purify HL and control the physicochemical properties of the soluble lignin fractions by simply altering the composition of the solvent mixture, and therefore, expanding the possibilities in valorisation of HL.
- 9 July 2025
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessFractionation of Lignin with Aqueous Organic Solvents: A Step Closer to Sustainable Wood Biorefinery
Recent rapid developments in forest biomass valorisation have highlighted lignin as a key value driver for the economic sustainability of modern biorefining. Sugar-based biorefinery hydrolysis lignin (HL) from wood biomass has an immense potential as an alternative for fossil-based aromatic feedstock. However, impurities, heterogeneous structures and limited solubility in common solvents are the main challenges associated with HL functionalisation. Aqueous organic solvent fractionation represents a promising approach to improve on HL properties and subsequently increase its valorisation options. In this study, for the first time, biorefinery HL derived from birch was purified and fractionated with water-EtOH and -THF mixtures across the whole co-solvent concentration range at ambient temperature. The highest lignin solubility was achieved with 70 wt% THF mixture, which dissolved a remarkable 81 wt% of lignin available in HL. Moreover, fractionated lignin purity was further increased to 95 wt% when employing an extra water-washing step. In general, fractions soluble in THF mixtures had high dispersity, syringyl-to-guaiacyl (S/G) subunit ratio and aliphatic OH-groups, whereas EtOH fractions had narrower molecular weight distributions, lower S/G subunit ratio and higher concentration of phenolic OH-groups. A brief description of the solvent effects at the molecular level was given using molecular dynamics simulations which revealed specific solvation between the dissolved lignin and solvent mixture. Protocol presented in this paper enables to purify HL and control the physicochemical properties of the soluble lignin fractions by simply altering the composition of the solvent mixture, and therefore, expanding the possibilities in valorisation of HL.
- 2 May 2025
- Published by Springer Nature
- p. 154-166
Counterintuitive Behavior of Clustering Quality: Findings for K-Means on Synthetic and Real Data
- 2 May 2025
- Published by Springer Nature
- p. 154-166
Counterintuitive Behavior of Clustering Quality: Findings for K-Means on Synthetic and Real Data
- 3 January 2025
- Published by Springer Nature in Pattern Analysis and Applications
- Vol. 28 (1) , 1-13
Learning Learning Curves
- 3 January 2025
- Published by Springer Nature in Pattern Analysis and Applications
- Vol. 28 (1) , 1-13
Learning Learning Curves
- Open Access
Inductive Conformal Prediction under Data Scarcity: Exploring the Impacts of Nonconformity Measures
Conformal prediction, which makes no distributional assumptions about the data, has emerged as a powerful and reliable approach to uncertainty quantification in practical applications. The nonconformity measure used in conformal prediction quantifies how a test sample differs from the training data and the effectiveness of a conformal prediction interval may depend heavily on the precise measure employed. The impact of this choice has, however, not been widely explored, especially when dealing with limited amounts of data. The primary objective of this study is to evaluate the performance of various nonconformity measures (absolute error-based, normalized absolute error-based, and quantile-based measures) in terms of validity and efficiency when used in inductive conformal prediction. The focus is on small datasets, which is still a common setting in many real-world applications. Using synthetic and real-world data, we assess how different characteristics -- such as dataset size, noise, and dimensionality -- can affect the efficiency of conformal prediction intervals. Our results show that although there are differences, no single nonconformity measure consistently outperforms the others, as the effectiveness of each nonconformity measure is heavily influenced by the specific nature of the data. Additionally, we found that increasing dataset size does not always improve efficiency, suggesting the importance of fine-tuning models and, again, the need to carefully select the nonconformity measure for different applications.
Open AccessInductive Conformal Prediction under Data Scarcity: Exploring the Impacts of Nonconformity Measures
Conformal prediction, which makes no distributional assumptions about the data, has emerged as a powerful and reliable approach to uncertainty quantification in practical applications. The nonconformity measure used in conformal prediction quantifies how a test sample differs from the training data and the effectiveness of a conformal prediction interval may depend heavily on the precise measure employed. The impact of this choice has, however, not been widely explored, especially when dealing with limited amounts of data. The primary objective of this study is to evaluate the performance of various nonconformity measures (absolute error-based, normalized absolute error-based, and quantile-based measures) in terms of validity and efficiency when used in inductive conformal prediction. The focus is on small datasets, which is still a common setting in many real-world applications. Using synthetic and real-world data, we assess how different characteristics -- such as dataset size, noise, and dimensionality -- can affect the efficiency of conformal prediction intervals. Our results show that although there are differences, no single nonconformity measure consistently outperforms the others, as the effectiveness of each nonconformity measure is heavily influenced by the specific nature of the data. Additionally, we found that increasing dataset size does not always improve efficiency, suggesting the importance of fine-tuning models and, again, the need to carefully select the nonconformity measure for different applications.
- Open Access
An Analysis of Model-Based Reinforcement Learning From Abstracted Observations
Many methods for Model-based Reinforcement learning (MBRL) in Markov decision processes (MDPs) provide guarantees for both the accuracy of the model they can deliver and the learning efficiency. At the same time, state abstraction techniques allow for a reduction of the size of an MDP while maintaining a bounded loss with respect to the original problem. Therefore, it may come as a surprise that no such guarantees are available when combining both techniques, i.e., where MBRL merely observes abstract states. Our theoretical analysis shows that abstraction can introduce a dependence between samples collected online (e.g., in the real world). That means that, without taking this dependence into account, results for MBRL do not directly extend to this setting. Our result shows that we can use concentration inequalities for martingales to overcome this problem. This result makes it possible to extend the guarantees of existing MBRL algorithms to the setting with abstraction. We illustrate this by combining R-MAX, a prototypical MBRL algorithm, with abstraction, thus producing the first performance guarantees for model-based 'RL from Abstracted Observations': model-based reinforcement learning with an abstract model.
Open AccessAn Analysis of Model-Based Reinforcement Learning From Abstracted Observations
Many methods for Model-based Reinforcement learning (MBRL) in Markov decision processes (MDPs) provide guarantees for both the accuracy of the model they can deliver and the learning efficiency. At the same time, state abstraction techniques allow for a reduction of the size of an MDP while maintaining a bounded loss with respect to the original problem. Therefore, it may come as a surprise that no such guarantees are available when combining both techniques, i.e., where MBRL merely observes abstract states. Our theoretical analysis shows that abstraction can introduce a dependence between samples collected online (e.g., in the real world). That means that, without taking this dependence into account, results for MBRL do not directly extend to this setting. Our result shows that we can use concentration inequalities for martingales to overcome this problem. This result makes it possible to extend the guarantees of existing MBRL algorithms to the setting with abstraction. We illustrate this by combining R-MAX, a prototypical MBRL algorithm, with abstraction, thus producing the first performance guarantees for model-based 'RL from Abstracted Observations': model-based reinforcement learning with an abstract model.
- 4 August 2023
- Published by Springer Nature
- p. 65-77
A View on Model Misspecification in Uncertainty Quantification
- 4 August 2023
- Published by Springer Nature
- p. 65-77
A View on Model Misspecification in Uncertainty Quantification
- 4 August 2023
- Published by Springer Nature
- p. 133-148
Model-Based Reinforcement Learning with State Abstraction: A Survey
- 4 August 2023
- Published by Springer Nature
- p. 133-148
Model-Based Reinforcement Learning with State Abstraction: A Survey