References of "Geurts, Pierre"
     in
Bookmark and Share    
Full Text
See detailUnsupervised gene network inference with decision trees and Random forests
Huynh-Thu, Vân Anh ULiege; Geurts, Pierre ULiege

in Sanguinetti, Guido; Huynh-Thu, Vân Anh (Eds.) Gene Regulatory Networks (2019)

In this chapter, we introduce the reader to a popular family of machine learning algorithms, called decision trees. We then review several approaches based on decision trees that have been developed for ... [more ▼]

In this chapter, we introduce the reader to a popular family of machine learning algorithms, called decision trees. We then review several approaches based on decision trees that have been developed for the inference of gene regulatory networks (GRNs). Decision trees have indeed several nice properties that make them well-suited for tackling this problem: they are able to detect multivariate interacting effects between variables, are non-parametric, have good scalability, and have very few parameters. In particular, we describe in detail the GENIE3 algorithm, a state-of-the-art method for GRN inference. [less ▲]

Detailed reference viewed: 71 (16 ULiège)
Full Text
Peer Reviewed
See detailDeep Quality Value (DQV) Learning
Sabatelli, Matthia ULiege; Louppe, Gilles ULiege; Geurts, Pierre ULiege et al

in Advances in Neural Information Processing Systems, Deep Reinforcement Learning Workshop (2018)

We introduce a novel Deep Reinforcement Learning (DRL) algorithm called Deep Quality-Value (DQV) Learning. DQV uses temporal-difference learning to train a Value neural network and uses this network for ... [more ▼]

We introduce a novel Deep Reinforcement Learning (DRL) algorithm called Deep Quality-Value (DQV) Learning. DQV uses temporal-difference learning to train a Value neural network and uses this network for training a second Quality-value network that learns to estimate state-action values. We first test DQV’s update rules with Multilayer Perceptrons as function approximators on two classic RL problems, and then extend DQV with the use of Deep Convolutional Neural Networks, ‘Experience Replay’ and ‘Target Neural Networks’ for tackling four games of the Atari Arcade Learning environment. Our results show that DQV learns significantly faster and better than Deep Q-Learning and Double Deep Q-Learning, suggesting that our algorithm can potentially be a better performing synchronous temporal difference algorithm than what is currently present in DRL. [less ▲]

Detailed reference viewed: 62 (17 ULiège)
Full Text
Peer Reviewed
See detailDeep Transfer Learning for Art Classification Problems
Sabatelli, Matthia ULiege; Kestemont, Mike; Daelemans, Walter et al

in European Conference on Computer Vision (ECCV), 4th Workshop on Computer VISion for ART Analysis (VISART IV) (2018)

In this paper we investigate whether Deep Convolutional Neural Net- works (DCNNs), which have obtained state of the art results on the ImageNet challenge, are able to perform equally well on three ... [more ▼]

In this paper we investigate whether Deep Convolutional Neural Net- works (DCNNs), which have obtained state of the art results on the ImageNet challenge, are able to perform equally well on three different art classification problems. In particular, we assess whether it is beneficial to fine tune the net- works instead of just using them as off the shelf feature extractors for a sepa- rately trained softmax classifier. Our experiments show how the first approach yields significantly better results and allows the DCNNs to develop new selective attention mechanisms over the images, which provide powerful insights about which pixel regions allow the networks successfully tackle the proposed classi- fication challenges. Furthermore, we also show how DCNNs, which have been fine tuned on a large artistic collection, outperform the same architectures which are pre-trained on the ImageNet dataset only, when it comes to the classification of heritage objects from a different dataset. [less ▲]

Detailed reference viewed: 62 (6 ULiège)
Full Text
Peer Reviewed
See detailRandom Forests based group importance scores and their statistical interpretation: application for Alzheimer’s disease
Wehenkel, Marie ULiege; Sutera, Antonio ULiege; Bastin, Christine ULiege et al

in Frontiers in Neuroscience (2018), 12

Machine learning approaches have been increasingly used in the neuroimaging field for the design of computer-aided diagnosis systems. In this paper, we focus on the ability of these methods to provide ... [more ▼]

Machine learning approaches have been increasingly used in the neuroimaging field for the design of computer-aided diagnosis systems. In this paper, we focus on the ability of these methods to provide interpretable information about the brain regions that are the most informative about the disease or condition of interest. In particular, we investigate the benefit of group-based, instead of voxel-based, analyses in the context of Random forests. Assuming a prior division of the voxels into non overlapping groups (defined by an atlas), we propose several procedures to derive group importances from individual voxel importances derived from random forests models. We then adapt several permutation schemes to turn group importance scores into more interpretable statistical scores that allow to determine the truly relevant groups in the importance rankings. The good behavior of these methods is first assessed on artificial datasets. Then, they are applied on our own dataset of FDG-PET scans to identify the brain regions involved in the prognosis of Alzheimer's disease. [less ▲]

Detailed reference viewed: 60 (17 ULiège)
Full Text
Peer Reviewed
See detailPhase Identification of Smart Meters by Clustering Voltage Measurements
Olivier, Frédéric ULiege; Sutera, Antonio ULiege; Geurts, Pierre ULiege et al

in Proceedings of the XX Power Systems Computation Conference (PSCC 2018) (2018, June)

When a smart meter, be it single-phase or threephase, is connected to a three-phase network, the phase(s) to which it is connected is (are) initially not known. This means that each of its measurements is ... [more ▼]

When a smart meter, be it single-phase or threephase, is connected to a three-phase network, the phase(s) to which it is connected is (are) initially not known. This means that each of its measurements is not uniquely associated with a phase of the distribution network. This phase information is important because it can be used by Distribution System Operators to take actions in order to have a network that is more balanced. In this work, the correlation between the voltage measurements of the smart meters is used to identify the phases. To do so, the constrained k-means clustering method is first introduced as a reference, as it has been previously used for phase identification. A novel, automatic and effective method is then proposed to overcome the main drawback of the constrained k-means clustering, and improve the quality of the clustering. Indeed, it takes into account the underlying structure of the low-voltage distribution networks beneath the voltage measurements without a priori knowledge on the topology of the network. Both methods are analysed with real measurements from a distribution network in Belgium. The proposed algorithm shows superior performance in different settings, e.g. when the ratio of single-phase over three- phase meters in the network is high, when the period over which the voltages are averaged is longer than one minute, etc. [less ▲]

Detailed reference viewed: 1028 (53 ULiège)
Full Text
Peer Reviewed
See detailComputer Aided Diagnosis System Based on Random Forests for the Prognosis of Alzheimer’s Disease
Wehenkel, Marie ULiege; Bastin, Christine ULiege; Geurts, Pierre ULiege et al

in 1st HBP Student Conference - Transdisciplinary Research Linking Neuroscience, Brain Medicine and Computer Science (2018, April)

In this abstract, we propose an original CAD system consisting in the combination of brain parcelling, ensemble of trees methods, and selection of (groups of) features using the importance scores embedded ... [more ▼]

In this abstract, we propose an original CAD system consisting in the combination of brain parcelling, ensemble of trees methods, and selection of (groups of) features using the importance scores embedded in tree-based methods. Indeed, on top of their ease of use and accuracy without ad hoc parameter tuning, tree ensemble methods such as random forests (RF) (Breiman, 2001) or extremely randomized trees (ET) (Geurts et al., 2006) provide interpretable results in the form of feature importance scores. We also compare the performance and interpretability of our proposed method to standard RF and ET approaches, without feature selection, and to multiple kernel learning (MKL). The latter was shown to be an efficient method notably capable of dealing with anatomically defined regions of the brain by the use of multiple kernels. [less ▲]

Detailed reference viewed: 38 (5 ULiège)
Full Text
Peer Reviewed
See detailGlobal multi-output decision trees for interaction prediction
Pliakos, Konstantinos; Geurts, Pierre ULiege; Vens, Celine

in Machine Learning (2018), 107

Interaction data are characterized by two sets of objects, each described by their own set of features. They are often modeled as networks and the values of interest are the possible interactions between ... [more ▼]

Interaction data are characterized by two sets of objects, each described by their own set of features. They are often modeled as networks and the values of interest are the possible interactions between two instances, represented usually as a matrix. Here, a novel global decision tree learning method is proposed, where multi-output decision trees are constructed over the global interaction setting, addressing the problem of interaction prediction as a multi-label classification task. More specifically, the tree is constructed bysplitting the interaction matrix both row-wise and column-wise, incorporating this way both interaction dataset features in the learning procedure. Experiments are conducted across several heterogeneous interaction datasets from the biomedical domain. The experimental results indicate the superiority of the proposed method against other decision tree approaches in terms of predictive accuracy, model size and computational efficiency. The performance is boosted by fully exploiting the multi-output structure of the model. We conclude that the proposed method should be considered in interaction prediction tasks, especially where interpretable models are desired. [less ▲]

Detailed reference viewed: 16 (1 ULiège)
Full Text
Peer Reviewed
See detailComparison of deep transfer learning strategies for digital pathology
Mormont, Romain ULiege; Geurts, Pierre ULiege; Marée, Raphaël ULiege

in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (2018)

In this paper, we study deep transfer learning as a way of overcoming object recognition challenges encountered in the field of digital pathology. Through several experiments, we investigate various uses ... [more ▼]

In this paper, we study deep transfer learning as a way of overcoming object recognition challenges encountered in the field of digital pathology. Through several experiments, we investigate various uses of pre-trained neural network architectures and different combination schemes with random forests for feature selection. Our experiments on eight classification datasets show that densely connected and residual networks consistently yield best performances across strategies. It also appears that network fine-tuning and using inner layers features are the best performing strategies, with the former yielding slightly superior results. [less ▲]

Detailed reference viewed: 503 (82 ULiège)
Full Text
Peer Reviewed
See detailLandmark detection in 2D bioimages for geometric morphometrics: a multi-resolution tree-based approach.
Vandaele, Rémy ULiege; Aceto, Jessica; Muller, Marc ULiege et al

in Scientific Reports (2018), 8(1), 538

The detection of anatomical landmarks in bioimages is a necessary but tedious step for geometric morphometrics studies in many research domains. We propose variants of a multi-resolution tree-based ... [more ▼]

The detection of anatomical landmarks in bioimages is a necessary but tedious step for geometric morphometrics studies in many research domains. We propose variants of a multi-resolution tree-based approach to speed-up the detection of landmarks in bioimages. We extensively evaluate our method variants on three different datasets (cephalometric, zebrafish, and drosophila images). We identify the key method parameters (notably the multi-resolution) and report results with respect to human ground truths and existing methods. Our method achieves recognition performances competitive with current existing approaches while being generic and fast. The algorithms are integrated in the open-source Cytomine software and we provide parameter configuration guidelines so that they can be easily exploited by end-users. Finally, datasets are readily available through a Cytomine server to foster future research. [less ▲]

Detailed reference viewed: 88 (15 ULiège)
Full Text
Peer Reviewed
See detailRandom Subspace with Trees for Feature Selection Under Memory Constraints
Sutera, Antonio ULiege; Châtel, Célia; Louppe, Gilles ULiege et al

in Storkey, Amos; Perez-Cruz, Fernando (Eds.) Proceedings of the Twenty-First International Conference on Artificial Intelligence and Statistics (2018)

Dealing with datasets of very high dimension is a major challenge in machine learning. In this paper, we consider the problem of feature selection in applications where the memory is not large enough to ... [more ▼]

Dealing with datasets of very high dimension is a major challenge in machine learning. In this paper, we consider the problem of feature selection in applications where the memory is not large enough to contain all features. In this setting, we propose a novel tree-based feature selection approach that builds a sequence of randomized trees on small subsamples of variables mixing both variables already identified as relevant by previous models and variables randomly selected among the other variables. As our main contribution, we provide an in-depth theoretical analysis of this method in infinite sample setting. In particular, we study its soundness with respect to common definitions of feature relevance and its convergence speed under various variable dependance scenarios. We also provide some preliminary empirical results highlighting the potential of the approach. [less ▲]

Detailed reference viewed: 50 (17 ULiège)
Full Text
See detailNets versus trees for feature ranking and gene network inference
Vecoven, Nicolas ULiege; Begon, Jean-Michel ULiege; Huynh-Thu, Vân Anh ULiege et al

E-print/Working paper (2017)

We propose to tackle the challenging problem of gene regulatory network inference, using variable importance measures derived from artifi cial neural networks (ANN). When combined with a L1-regularized ... [more ▼]

We propose to tackle the challenging problem of gene regulatory network inference, using variable importance measures derived from artifi cial neural networks (ANN). When combined with a L1-regularized selection layer, these measures allow ANN to be competitive with state of the art techniques for this problem based on random forests. [less ▲]

Detailed reference viewed: 56 (25 ULiège)
Full Text
Peer Reviewed
See detailSepsis prediction in critically ill patients by platelet activation markers on ICU admission: a prospective pilot study
LAYIOS, Nathalie ULiege; Delierneux, Céline ULiege; Hego, Alexandre ULiege et al

in Intensive Care Medicine Experimental (2017), 5(1), 32

Background: Platelets have been involved in both surveillance and host defense against severe infection. To date, whether platelet phenotype or other hemostasis components could be associated with ... [more ▼]

Background: Platelets have been involved in both surveillance and host defense against severe infection. To date, whether platelet phenotype or other hemostasis components could be associated with predisposition to sepsis in critical illness remains unknown. The aim of this work was to identify platelet markers that could predict sepsis occurrence in critically ill injured patients. Results: This single-center, prospective, observational, 7-month study was based on a cohort of 99 non-infected adult patients admitted to ICUs for elective cardiac surgery, trauma, acute brain injury and post-operative prolonged ventilation and followed up during ICU stay. Clinical characteristics and severity score (SOFA) were recorded on admission. Platelet activation markers, including fibrinogen binding to platelets, platelet membrane P-selectin expression, plasma soluble CD40L, and platelet-leukocytes aggregates were assayed by flow cytometry at admission and 48h later, and also at the time of sepsis diagnosis (Sepsis-3 criteria) and 7 days later for sepsis patients. Hospitalization data and outcomes were also recorded. Of the 99 patients, 19 developed sepsis after a median time of 5 days. SOFA at admission was higher; their levels of fibrinogen binding to platelets (platelet-Fg) and of D-dimers were significantly increased compared to the other patients. Levels 48h after ICU admission were no longer significant. Platelet-Fg % was an independent predictor of sepsis (P = 0.030). By ROC curve analysis cutoff points for SOFA (AUC=0.85) and Platelet-Fg (AUC=0.75) were 8 and 50%, respectively. The prior risk of sepsis (19%) increased to 50% when SOFA was above 8, to 46% when Platelet-Fg was above 50%, and to 87% when both SOFA and Platelet-Fg were above their cutoff values. By contrast, when the two parameters were below their cutoffs, the risk of sepsis was negligible (3.8%). Patients with sepsis had longer ICU and hospital stays and higher death rate. Conclusion: In addition to SOFA, platelet-bound fibrinogen levels assayed by flow cytometry within 24h of ICU admission help identifying critically ill patients at risk of developing sepsis. [less ▲]

Detailed reference viewed: 130 (32 ULiège)
Full Text
Peer Reviewed
See detailSCENIC: single-cell regulatory network inference and clustering
Aibar, Sara; González-Blas, Carmen Bravo; Moerman, Thomas et al

in Nature Methods (2017), 14

We present SCENIC, a computational method for simultaneous gene regulatory network reconstruction and cell-state identification from single-cell RNA-seq data (http://scenic.aertslab.org). On a compendium ... [more ▼]

We present SCENIC, a computational method for simultaneous gene regulatory network reconstruction and cell-state identification from single-cell RNA-seq data (http://scenic.aertslab.org). On a compendium of single-cell data from tumors and brain, we demonstrate that cis-regulatory analysis can be exploited to guide the identification of transcription factors and cell states. SCENIC provides critical biological insights into the mechanisms driving cellular heterogeneity. [less ▲]

Detailed reference viewed: 170 (9 ULiège)
Full Text
Peer Reviewed
See detailA two-step methodology for human pose estimation increasing the accuracy and reducing the amount of learning samples dramatically
Azrour, Samir ULiege; Pierard, Sébastien ULiege; Geurts, Pierre ULiege et al

in Advanced Concepts for Intelligent Vision Systems (2017, September)

In this paper, we present a two-step methodology to improve existing human pose estimation methods from a single depth image. Instead of learning the direct mapping from the depth image to the 3D pose, we ... [more ▼]

In this paper, we present a two-step methodology to improve existing human pose estimation methods from a single depth image. Instead of learning the direct mapping from the depth image to the 3D pose, we first estimate the orientation of the standing person seen by the camera and then use this information to dynamically select a pose estimation model suited for this particular orientation. We evaluated our method on a public dataset of realistic depth images with precise ground truth joints location. Our experiments show that our method decreases the error of a state-of-the-art pose estimation method by 30%, or reduces the size of the needed learning set by a factor larger than 10. [less ▲]

Detailed reference viewed: 92 (14 ULiège)
Full Text
Peer Reviewed
See detailTree Ensemble Methods and Parcelling to Identify Brain Areas Related to Alzheimer’s Disease
Wehenkel, Marie ULiege; Bastin, Christine ULiege; Phillips, Christophe ULiege et al

in 2017 International Workshop on Pattern Recognition in Neuroimaging (PRNI), proceedings (2017, June)

Detailed reference viewed: 93 (22 ULiège)
Full Text
Peer Reviewed
See detailSimple connectome inference from partial correlation statistics in calcium imaging
Sutera, Antonio ULiege; Joly, Arnaud ULiege; François-Lavet, Vincent et al

in Soriano, Jordi; Battaglia, Demian; Guyon, Isabelle (Eds.) et al Neural Connectomics Challenge (2017)

In this work, we propose a simple yet effective solution to the problem of connectome inference in calcium imaging data. The proposed algorithm consists of two steps. First, processing the raw signals to ... [more ▼]

In this work, we propose a simple yet effective solution to the problem of connectome inference in calcium imaging data. The proposed algorithm consists of two steps. First, processing the raw signals to detect neural peak activities. Second, inferring the degree of association between neurons from partial correlation statistics. This paper summarises the methodology that led us to win the Connectomics Challenge, proposes a simplified version of our method, and finally compares our results with respect to other inference methods. [less ▲]

Detailed reference viewed: 155 (15 ULiège)
Full Text
Peer Reviewed
See detailGlobally Induced Forest: A Prepruning Compression Scheme
Begon, Jean-Michel ULiege; Joly, Arnaud; Geurts, Pierre ULiege

in Proceedings of Machine Learning Research (2017), 70

Tree-based ensemble models are heavy memory- wise. An undesired state of affairs consider- ing nowadays datasets, memory-constrained environment and fitting/prediction times. In this paper, we propose the ... [more ▼]

Tree-based ensemble models are heavy memory- wise. An undesired state of affairs consider- ing nowadays datasets, memory-constrained environment and fitting/prediction times. In this paper, we propose the Globally Induced Forest (GIF) to remedy this problem. GIF is a fast prepruning approach to build lightweight ensembles by iteratively deepening the current forest. It mixes local and global optimizations to produce accurate predictions under memory constraints in reasonable time. We show that the proposed method is more than competitive with standard tree-based ensembles under corresponding constraints, and can sometimes even surpass much larger models. [less ▲]

Detailed reference viewed: 152 (35 ULiège)