Browsing
     by title


0-9 A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

or enter first few letters:   
OK
See detailExploitation pédagogique de la télédétection par satellite
Binard, Marc ULg; Marchal, Denis; Donnay, Jean-Paul ULg

in Information et Enseignement (IBM) (1989), 15

Detailed reference viewed: 15 (1 ULg)
See detailExploitations philologiques et historiques d’une banque de données de la langue latine
Purnelle, Gérald ULg

in Cacaly, Serge; Losfeld, Gérard (Eds.) Sciences historiques, sciences du passé et nouvelles technologies d'information. Bilan et évaluation (1989)

Detailed reference viewed: 17 (1 ULg)
Full Text
See detailExploiter des scènes de lecture et des fictions métanarratives pour accompagner les élèves en difficulté au début de l'enseignement secondaire
De Croix, Séverine ULg

in Dufays, Jean-Louis (Ed.) Enseigner et apprendre la littérature aujourd'hui, pour quoi faire ? Sens, utilité, évaluation (2007)

Detailed reference viewed: 29 (1 ULg)
See detailExploiter et gérer durablement les forêts d'Afrique Centrale
Doucet, Jean-Louis ULg

Speech/Talk (2009)

Detailed reference viewed: 35 (10 ULg)
See detailExploiter les méthodes de gestion des risques de marché et la VaR comme interface à la gestion d'actifs
Esch, Louis ULg; Lopez, Thierry

Scientific conference (2000, December)

Detailed reference viewed: 5 (0 ULg)
See detailExploiting Biology Specific Properties for the Estimation of Kinetic Parameters
Fey, D.; Bullinger, Eric ULg

Poster (2008, August)

Detailed reference viewed: 7 (0 ULg)
Full Text
Peer Reviewed
See detailExploiting Electron Spectroscopies to Probe the Structure and Organization of Self-assembled Monolayers : a Review
Duwez, Anne-Sophie ULg

in Journal of Electron Spectroscopy & Related Phenomena (2004), 134

In this paper, we review and discuss the use of electron spectroscopies to characterize self-assembled monolayers (SAMs). The review concentrates on thiol-derived monolayers on gold with emphasis on n ... [more ▼]

In this paper, we review and discuss the use of electron spectroscopies to characterize self-assembled monolayers (SAMs). The review concentrates on thiol-derived monolayers on gold with emphasis on n-alkanethiols, considered as archetypal systems. Since they are relatively simple model systems (ease of preparation, high structural order, and flexibility in the structure of chemical groups exposed to the surface), they are particularly well-suited reference samples for molecular level understanding of surface phenomena and for disclosing the potential of surface sensitive techniques. Some examples concerning silane-derived monolayers are also discussed. Many different spectroscopic techniques have been applied to characterize SAMs. Among them, electron spectroscopies, such as XPS, UPS, photoemission with synchrotron radiation, and HREELS, have been used to investigate the structure of alkanethiols on gold, and in particular, to characterize the S–Au bond, the packing density, the crystalline order, and molecular orientation.We try here to provide an overview on the structural information that can be obtained from those techniques. Damage processes induced by X-ray and electron beam are discussed in detail. [less ▲]

Detailed reference viewed: 30 (0 ULg)
Full Text
See detailExploiting Interspecific Genetic Variability for Improving Common bean for higher productivity on soils presenting biotic and abiotic stresses
Butare, Louis ULg

Doctoral thesis (2015)

Biotic and abiotic stresses often occur in the same field of rural poor farmer households in tropical countries resulting in heavy losses of common bean yields. To improve resistance of common beans ... [more ▼]

Biotic and abiotic stresses often occur in the same field of rural poor farmer households in tropical countries resulting in heavy losses of common bean yields. To improve resistance of common beans, sensitive Phaseolus vulgaris (SER16) was crossed to resistant P. coccineus (G35346-3Q) to create 94 F5:6 recombinant inbred lines (RILs) of the pedigree SER16♀ x (SER16♀ x G35346-3Q♂). The objectives of this study were to (i) identify potential parents for resistance to Al, drought and Fusarium root rot among 11 bean genotypes, (ii) to evaluate 94 F5:6 Recombinant Inbred Lines (RILs) of the cross SER 16♀ x (SER 16♀ x G35346-3Q♂) both for their resistance to Al and /or drought, (iii) to evaluate RILs for resistance to Fusarium root rot, and (iv) to identify QTL for resistance to these stresses. RILs were characterized in greenhouse for resistance to Al using a hydroponic screening employing a nutrient solution with or without 20 μM Al , to Al-toxic acid soil with high Al (HAl) and low Al (LAl) saturation, to terminal drought simulation with and without progressive soil drying, to combined stresses of Al and terminal drought in 80 cm long soil cylinder system, and to Fusarium root rot using inoculated perlite soil and sand (2:1). Two field studies were also carried on in Colombia under rainfed and irrigated conditions in Palmira, and high Al saturated acid soil in Santander of Quilichao. Our studies confirmed the superiority in Al response of Andean common beans in greenhouse trials compared to Middle American type for several root traits. Each screening method of our Al greenhouse experiments permitted an evaluation of different aspects of root traits. The two parents were virtually equal for tap root elongation rate at 24 h in the 20 μM Al treatment at about 1.4 mm h-1 while progenies ranged from less than 1-1.75 mm h-1. The correlation between leaf area and total root length was highly significant under high Al saturation (r = 0.70***) for HAl-acid soil. Two genotypes (ALB88 and ALB 91) emerged as strong multiple trait lines for the two abiotic stresses. Fusarium root rot induced root growth inhibition as high as 80.8% for the susceptible ALB 5, while resistant RILs (ALB45, ALB41, ALB126, ALB84, ALB49, ALB34, ALB88 and ALB85) didn`t show any inhibition . Seed yield under drought stress conditions was positively associated to 100-seed weight both under irrigated field (r = 0.28**) and rainfed field (r = 0.36***), and negatively associated to days to maturity (DTM) (r = - 0.36***) in field evaluation in Al-toxic acid soil in Quilichao (Colombia). QTLs for important traits including root characteristics under high Al , grain yield and yield components for drought and high Al saturation soil were identified. The use of both soil and hydroponic system, and field could contribute to evaluation of breeding materials to identify genotypes that combine Al resistance with acid soil tolerance, drought and root rot tolerance. [less ▲]

Detailed reference viewed: 235 (14 ULg)
Full Text
Peer Reviewed
See detailExploiting Localization for Faster Power System Dynamic Simulations
Aristidou, Petros ULg; Fabozzi, Davide; Van Cutsem, Thierry ULg

in Proc. IEEE PES 2013 PowerTech Conference (2013, June)

This paper proposes an algorithm for exploiting the localized response of power system components to accelerate dynamic simulations. During the simulation, components marginally participating to the ... [more ▼]

This paper proposes an algorithm for exploiting the localized response of power system components to accelerate dynamic simulations. During the simulation, components marginally participating to the system dynamics are characterized as latent and their dynamic models are replaced by much simpler equivalents. At the same time, components with significant dynamic activity are characterized as active and their original dynamic models are used. Based on the criterion proposed, components switch status between active and latent to increase performance while retaining accuracy. Two realistic test systems, a medium-scale and a large-scale, are used for the performance evaluation of the proposed method. [less ▲]

Detailed reference viewed: 187 (39 ULg)
Full Text
Peer Reviewed
See detailExploiting policy knowledge in online least-squares policy iteration: An empirical study
Busoniu, Lucian; Ernst, Damien ULg; Babusku, Robert et al

in Automation, Computers, Applied Mathematics (2010), 19(4), 521-529

Reinforcement learning (RL) is a promising paradigm for learning optimal control. Traditional RL works for discrete variables only, so to deal with the continuous variables appearing in control problems ... [more ▼]

Reinforcement learning (RL) is a promising paradigm for learning optimal control. Traditional RL works for discrete variables only, so to deal with the continuous variables appearing in control problems, approximate representations of the solution are necessary. The field of approximate RL has tremendously expanded over the last decade, and a wide array of effective algorithms is now available. However, RL is generally envisioned as working without any prior knowledge about the system or the solution, whereas such knowledge is often available and can be exploited to great advantage. Therefore, in this paper we describe a method that exploits prior knowledge to accelerate online least-squares policy iteration (LSPI), a state-of-the-art algorithm for approximate RL. We focus on prior knowledge about the monotonicity of the control policy with respect to the system states. Such monotonic policies are appropriate for important classes of systems appearing in control applications, including for instance nearly linear systems and linear systems with monotonic input nonlinearities. In an empirical evaluation, online LSPI with prior knowledge is shown to learn much faster and more reliably than the original online LSPI. [less ▲]

Detailed reference viewed: 87 (4 ULg)
Full Text
See detailExploiting random projections and sparsity with random forests and gradient boosting methods - Application to multi-label and multi-output learning, random forest model compression and leveraging input sparsity
Joly, Arnaud ULg

Doctoral thesis (2017)

Within machine learning, the supervised learning field aims at modeling the input-output relationship of a system, from past observations of its behavior. Decision trees characterize the input-output ... [more ▼]

Within machine learning, the supervised learning field aims at modeling the input-output relationship of a system, from past observations of its behavior. Decision trees characterize the input-output relationship through a series of nested ``if-then-else'' questions, the testing nodes, leading to a set of predictions, the leaf nodes. Several of such trees are often combined together for state-of-the-art performance: random forest ensembles average the predictions of randomized decision trees trained independently in parallel, while tree boosting ensembles train decision trees sequentially to refine the predictions made by the previous ones. The emergence of new applications requires scalable supervised learning algorithms in terms of computational power and memory space with respect to the number of inputs, outputs, and observations without sacrificing accuracy. In this thesis, we identify three main areas where decision tree methods could be improved for which we provide and evaluate original algorithmic solutions: (i) learning over high dimensional output spaces, (ii) learning with large sample datasets and stringent memory constraints at prediction time and (iii) learning over high dimensional sparse input spaces. A first approach to solve learning tasks with a high dimensional output space, called binary relevance or single target, is to train one decision tree ensemble per output. However, it completely neglects the potential correlations existing between the outputs. An alternative approach called multi-output decision trees fits a single decision tree ensemble targeting simultaneously all the outputs, assuming that all outputs are correlated. Nevertheless, both approaches have (i) exactly the same computational complexity and (ii) target extreme output correlation structures. In our first contribution, we show how to combine random projection of the output space, a dimensionality reduction method, with the random forest algorithm decreasing the learning time complexity. The accuracy is preserved, and may even be improved by reaching a different bias-variance tradeoff. In our second contribution, we first formally adapt the gradient boosting ensemble method to multi-output supervised learning tasks such as multi-output regression and multi-label classification. We then propose to combine single random projections of the output space with gradient boosting on such tasks to adapt automatically to the output correlation structure. The random forest algorithm often generates large ensembles of complex models thanks to the availability of a large number of observations. However, the space complexity of such models, proportional to their total number of nodes, is often prohibitive, and therefore these modes are not well suited under stringent memory constraints at prediction time. In our third contribution, we propose to compress these ensembles by solving a L1-based regularization problem over the set of indicator functions defined by all their nodes. Some supervised learning tasks have a high dimensional but sparse input space, where each observation has only a few of the input variables that have non zero values. Standard decision tree implementations are not well adapted to treat sparse input spaces, unlike other supervised learning techniques such as support vector machines or linear models. In our fourth contribution, we show how to exploit algorithmically the input space sparsity within decision tree methods. Our implementation yields a significant speed up both on synthetic and real datasets, while leading to exactly the same model. It also reduces the required memory to grow such models by exploiting sparse instead of dense memory storage for the input matrix. [less ▲]

Detailed reference viewed: 150 (9 ULg)
Full Text
Peer Reviewed
See detailExploiting semi-analytical sensitivities from linear and non-linear finite element analyses for composite panel optimisation
Bruyneel, Michaël ULg; Coslon, Benoit; Delsemme, Jean-Pierre et al

in International Journal of Structural Stability & Dynamics (2010), 10

Detailed reference viewed: 16 (1 ULg)
Full Text
Peer Reviewed
See detailExploiting SNP Correlations within Random Forest for Genome-Wide Association Studies
Botta, Vincent ULg; Louppe, Gilles ULg; Geurts, Pierre ULg et al

in PLoS ONE (2014)

The primary goal of genome-wide association studies (GWAS) is to discover variants that could lead, in isolation or in combination, to a particular trait or disease. Standard approaches to GWAS, however ... [more ▼]

The primary goal of genome-wide association studies (GWAS) is to discover variants that could lead, in isolation or in combination, to a particular trait or disease. Standard approaches to GWAS, however, are usually based on univariate hypothesis tests and therefore can account neither for correlations due to linkage disequilibrium nor for combinations of several markers. To discover and leverage such potential multivariate interactions, we propose in this work an extension of the Random Forest algorithm tailored for structured GWAS data. In terms of risk prediction, we show empirically on several GWAS datasets that the proposed T-Trees method significantly outperforms both the original Random Forest algorithm and standard linear models, thereby suggesting the actual existence of multivariate non-linear effects due to the combinations of several SNPs. We also demonstrate that variable importances as derived from our method can help identify relevant loci. Finally, we highlight the strong impact that quality control procedures may have, both in terms of predictive power and loci identification. [less ▲]

Detailed reference viewed: 219 (18 ULg)
Full Text
Peer Reviewed
See detailExploiting Socially-Generated Side Information in Dimensionality Reduction
Marcos Alvarez, Alejandro ULg; Yamada, Makoto; Kimura, Akisato

in Proceedings of the 2nd International Workshop on Socially-Aware Multimedia (2013, October)

In this paper, we show how side information extracted from socially-curated data can be used within a dimensionality reduction method and to what extent this side information is beneficial to several ... [more ▼]

In this paper, we show how side information extracted from socially-curated data can be used within a dimensionality reduction method and to what extent this side information is beneficial to several tasks such as image classification, data visualization and image retrieval. The key idea is to incorporate side information of an image into a dimensionality reduction method. More specifically, we propose a dimensionality reduction method that can find an embedding transformation so that images with similar side information are close in the embedding space. We introduce three types of side information derived from user behavior. Through experiments on images from Pinterest, we show that incorporating socially-generated side information in a dimensionality reduction method benefits several image-related tasks such as image classification, data visualization and image retrieval. [less ▲]

Detailed reference viewed: 64 (20 ULg)
Full Text
Peer Reviewed
See detailExploiting the Burkholderia pseudomallei acute phase antigen BPSL2765 for structure-based epitope discovery/design in structural vaccinology.
Gourlay, Louise J.; Peri, Claudio; Ferrer-Navarro, Mario et al

in Chemistry & biology (2013), 20(9), 1147-56

We solved the crystal structure of Burkholderia pseudomallei acute phase antigen BPSL2765 in the context of a structural vaccinology study, in the area of melioidosis vaccine development. Based on the ... [more ▼]

We solved the crystal structure of Burkholderia pseudomallei acute phase antigen BPSL2765 in the context of a structural vaccinology study, in the area of melioidosis vaccine development. Based on the structure, we applied a recently developed method for epitope design that combines computational epitope predictions with in vitro mapping experiments and successfully identified a consensus sequence within the antigen that, when engineered as a synthetic peptide, was selectively immunorecognized to the same extent as the recombinant protein in sera from melioidosis-affected subjects. Antibodies raised against the consensus peptide were successfully tested in opsonization bacterial killing experiments and antibody-dependent agglutination tests of B. pseudomallei. Our strategy represents a step in the development of immunodiagnostics, in the production of specific antibodies and in the optimization of antigens for vaccine development, starting from structural and physicochemical principles. [less ▲]

Detailed reference viewed: 44 (0 ULg)
Full Text
Peer Reviewed
See detailExploiting the combination of natural and genetically engineered resistance to cassava mosaic and cassava brown streak viruses impacting cassava production in Africa.
Vanderschuren, Hervé ULg; Moreno, Isabel; Anjanappa, Ravi B. et al

in PloS one (2012), 7(9), 45277

Cassava brown streak disease (CBSD) and cassava mosaic disease (CMD) are currently two major viral diseases that severely reduce cassava production in large areas of Sub-Saharan Africa. Natural resistance ... [more ▼]

Cassava brown streak disease (CBSD) and cassava mosaic disease (CMD) are currently two major viral diseases that severely reduce cassava production in large areas of Sub-Saharan Africa. Natural resistance has so far only been reported for CMD in cassava. CBSD is caused by two virus species, Cassava brown streak virus (CBSV) and Ugandan cassava brown streak virus (UCBSV). A sequence of the CBSV coat protein (CP) highly conserved between the two virus species was used to demonstrate that a CBSV-CP hairpin construct sufficed to generate immunity against both viral species in the cassava model cultivar (cv. 60444). Most of the transgenic lines showed high levels of resistance under increasing viral loads using a stringent top-grafting method of inoculation. No viral replication was observed in the resistant transgenic lines and they remained free of typical CBSD root symptoms 7 month post-infection. To generate transgenic cassava lines combining resistance to both CBSD and CMD the hairpin construct was transferred to a CMD-resistant farmer-preferred Nigerian landrace TME 7 (Oko-Iyawo). An adapted protocol allowed the efficient Agrobacterium-based transformation of TME 7 and the regeneration of transgenic lines with high levels of CBSV-CP hairpin-derived small RNAs. All transgenic TME 7 lines were immune to both CBSV and UCBSV infections. Further evaluation of the transgenic TME 7 lines revealed that CBSD resistance was maintained when plants were co-inoculated with East African cassava mosaic virus (EACMV), a geminivirus causing CMD. The innovative combination of natural and engineered virus resistance in farmer-preferred landraces will be particularly important to reducing the increasing impact of cassava viral diseases in Africa. [less ▲]

Detailed reference viewed: 21 (1 ULg)
Full Text
Peer Reviewed
See detailExploiting the use of DC SCOPF approximation to improve iterative AC SCOPF algorithms
Marano Marcolini, Alejandro; Capitanescu, Florin ULg; Jose Luis, Martinez Ramos et al

in IEEE Transactions on Power Systems (2012), 27(3), 1459-1466

This paper focuses on improving the solution techniques for the AC SCOPF problem of active power dispatch by using the DC SCOPF approximation within the SCOPF algorithm. Our approach brings two benefits ... [more ▼]

This paper focuses on improving the solution techniques for the AC SCOPF problem of active power dispatch by using the DC SCOPF approximation within the SCOPF algorithm. Our approach brings two benefits compared to benchmark SCOPF algorithms: it speeds-up the solution of an iterative AC SCOPF algorithm thanks to a more efficient identification of binding contingencies, and allows improving the objective by an appropriate choice of a limited number of corrective actions for each contingency. The proposed approach is illustrated on 5 test systems of 60, 118, 300, 1203, and 2746 buses. [less ▲]

Detailed reference viewed: 123 (9 ULg)