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
Full Text
Peer Reviewed
See detailFuzzy approximation for convergent model-based reinforcement learning
Busoniu, Lucian; Ernst, Damien ULg; Babuska, Robert et al

in Proceedings of the 2007 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE-07) (2007)

Reinforcement learning (RL) is a learning control paradigm that provides well-understood algorithms with good convergence and consistency properties. Unfortunately, these algorithms require that process ... [more ▼]

Reinforcement learning (RL) is a learning control paradigm that provides well-understood algorithms with good convergence and consistency properties. Unfortunately, these algorithms require that process states and control actions take only discrete values. Approximate solutions using fuzzy representations have been proposed in the literature for the case when the states and possibly the actions are continuous. However, the link between these mainly heuristic solutions and the larger body of work on approximate RL, including convergence results, has not been made explicit. In this paper, we propose a fuzzy approximation structure for the Q-value iteration algorithm, and show that the resulting algorithm is convergent. The proof is based on an extension of previous results in approximate RL. We then propose a modi ed, serial version of the algorithm that is guaranteed to converge at least as fast as the original algorithm. An illustrative simulation example is also provided. [less ▲]

Detailed reference viewed: 16 (1 ULg)
Full Text
Peer Reviewed
See detailFuzzy Control Chart A Better Alternative for Shewhart Average Chart
Faraz, Alireza ULg; MOGHADAM, M. B.

in Quality & Quantity (2007), 41(3), 375-385

This paper through a real illustrative example and a power test shows that designing a fuzzy control chart for process average of a continuous (variable) quality characteristic with a warning line is a ... [more ▼]

This paper through a real illustrative example and a power test shows that designing a fuzzy control chart for process average of a continuous (variable) quality characteristic with a warning line is a better alternative to Shewhart ¯X chart in many respects, like providing better neural view to inspectors, offering different strategic options for company to choose, detecting the desire shifts more quickly, and more sensibility to small shifts without any complexity augmentation to the chart. [less ▲]

Detailed reference viewed: 55 (10 ULg)
Full Text
Peer Reviewed
See detailFuzzy K-NN applied to moulds detection
Kuske, Martyna; Rubio, R.; Nicolas, Jacques ULg et al

in Proceedings of ISOEN'03 (2003, June)

The possibility to detect Aspergillus versicolor growing on different building materials by a metal oxide sensor array is studied. Results show that an accurate classification rate of 89 +/- 3% can be ... [more ▼]

The possibility to detect Aspergillus versicolor growing on different building materials by a metal oxide sensor array is studied. Results show that an accurate classification rate of 89 +/- 3% can be obtained combining an extended linear discriminant analysis plus a fuzzy k-NN classifier. The classification ability of the classifier is assessed within the dataset by crossvalidation and also in a second dataset collected 5 months later. There is a slight decrease in the classification performance for all the algorithms, being the most sensitive the most accurate one. [less ▲]

Detailed reference viewed: 21 (4 ULg)
Full Text
Peer Reviewed
See detailFuzzy k-NN applied to moulds detection
Kuske, Martyna; Rubio, R.; Romain, Anne-Claude ULg et al

in Sensors and actuators. B, Chemical (2005), 106(1), 52-60

The possibility to detect Aspergillus versicolor growing on different building materials by a metal oxide sensor array is studied. Results show that an accurate classification rate of 89 +/- 3% can be ... [more ▼]

The possibility to detect Aspergillus versicolor growing on different building materials by a metal oxide sensor array is studied. Results show that an accurate classification rate of 89 +/- 3% can be obtained combining an extended linear discriminant analysis plus a fuzzy k-NN classifier. The classification ability of the classifier is assessed within the dataset by crossvalidation and also in a second dataset collected 5 months later. There is a slight decrease in the classification performance for all the algorithms, being the most sensitive the most accurate one. [less ▲]

Detailed reference viewed: 69 (40 ULg)
Full Text
Peer Reviewed
See detailA Fuzzy Metric for Assessing the Producibility of Straightening in Early Design
Caprace, Jean-David ULg; Aracil Fernandez, Francisco; Losseau, Nicolas ULg et al

in The 14th International Conference on Computer Applications in Shipbuilding (ICCAS) (2009, September 01)

Since several years, the large shipyards use more and more thin plates to build up stiffened panels in order to decrease the structural weight. The major problem relating to the utilization of thin plates ... [more ▼]

Since several years, the large shipyards use more and more thin plates to build up stiffened panels in order to decrease the structural weight. The major problem relating to the utilization of thin plates is the occurrence of welding distortions that have to be eliminated for esthetical and service reasons. This straightening operation involves significant costs and it seems thus important to characterize its economical impact on the global hull fabrication. This paper presents a way to minimize cost in shipbuilding industry by the implementation of a producibility fuzzy metric in order to obtain a better knowledge of the straightening process. Attention has been focused on the definition of fuzzy rules by experts but also on the optimisation of the decision surface to reduce the error compared to actual measurement points. [less ▲]

Detailed reference viewed: 95 (16 ULg)
Full Text
Peer Reviewed
See detailFuzzy partition optimization for approximate fuzzy Q-iteration
Busoniu, Lucian; Ernst, Damien ULg; Babuska, Robert et al

in Proceedings of the 17th IFAC World Congress (IFAC-08) (2008)

Reinforcement learning (RL) is a widely used learning paradigm for adaptive agents. Because exact RL can only be applied to very simple problems, approximate algorithms are usually necessary in practice ... [more ▼]

Reinforcement learning (RL) is a widely used learning paradigm for adaptive agents. Because exact RL can only be applied to very simple problems, approximate algorithms are usually necessary in practice. Many algorithms for approximate RL rely on basis-function representations of the value function (or of the Q-function). Designing a good set of basis functions without any prior knowledge of the value function (or of the Q-function) can be a difficult task. In this paper, we propose instead a technique to optimize the shape of a constant number of basis functions for the approximate, fuzzy Q-iteration algorithm. In contrast to other approaches to adapt basis functions for RL, our optimization criterion measures the actual performance of the computed policies in the task, using simulation from a representative set of initial states. A complete algorithm, using cross-entropy optimization of triangular fuzzy membership functions, is given and applied to the car-on-the-hill example. [less ▲]

Detailed reference viewed: 83 (2 ULg)
Full Text
Peer Reviewed
See detailG protein-coupled receptors, an unexploited animal toxin targets: Exploration of green mamba venom for novel drug candidates active against adrenoceptors
Maïga, Arhamatoulaye; Mourier, Gilles; Quinton, Loïc ULg et al

in Toxicon (2012), 59

At a time when pharmaceutical companies are having trouble finding new low MW drugs and when biologics are becoming more common, animal venoms could constitute an underexploited source of novel drug ... [more ▼]

At a time when pharmaceutical companies are having trouble finding new low MW drugs and when biologics are becoming more common, animal venoms could constitute an underexploited source of novel drug candidates. We looked for identifying novel animal toxins active against G protein-coupled receptors (GPCR), the most frequently exploited class of treatment targets, with the aim to develop novel research tools and drug candidates. Screening of green mamba (Dendroaspis angusticeps) venom against adrenoceptors identified two novel venom peptides. r-Da1a shown an affinity of 0.35 nM for the a1a-AR while r-Da1b displayed affinities between 14 and 73 nM for the three a2-ARs. These two venom peptides have sequences similar to those of muscarinic toxins and belong to the three-finger-fold protein family. a1a-AR is the primary target for the treatment of prostate hypertrophy. In vitro and in vivo tests demonstrated that r-Da1a reduced prostatic muscle tone as efficiently as tamsulosin (an antagonist presently used), but with fewer cardiovascular side effects. a2-ARs are the prototype of GPCRs not currently used as treatment targets due to a lack of specific ligands. Blockage of these receptors increases intestinal motility, which may be compromised by abdominal surgery and reduces orthosteric hypotension. In vitro and in vivo tests demonstrated that r-Da1b antagonizes a2-ARs in smooth muscles and increased heart rate and blood catecholamine concentrations. These results highlight possible exploitation of r-Da1a and r-Da1b in important pathologies. [less ▲]

Detailed reference viewed: 14 (1 ULg)
See detail(G)astronomie - Barres météoritiques
Nazé, Yaël ULg

Article for general public (2013)

Detailed reference viewed: 2 (0 ULg)
See detail(G)astronomie - Big bang cookies
Nazé, Yaël ULg

Article for general public (2013)

Detailed reference viewed: 9 (0 ULg)
See detail(G)astronomie - Comète glacée
Nazé, Yaël ULg

Article for general public (2012)

Detailed reference viewed: 15 (2 ULg)
See detail(G)astronomie - Dip martien
Nazé, Yaël ULg

Article for general public (2013)

See detail(G)astronomie - Smoothie interstellaire
Nazé, Yaël ULg

Article for general public (2013)

Detailed reference viewed: 5 (0 ULg)
See detail(G)astronomie - Soufflé structurel
Nazé, Yaël ULg

Article for general public (2013)

Detailed reference viewed: 3 (1 ULg)
See detail(G)astronomie - Stoemp céleste
Nazé, Yaël ULg

Article for general public (2011)

Detailed reference viewed: 15 (4 ULg)
See detail(G)astronomie - Tarte Titan
Nazé, Yaël ULg

Article for general public (2011)

Detailed reference viewed: 15 (1 ULg)
Full Text
See detailG-Cube 2.0 : Un logiciel de cubage et de constitution de lots de bois
Handerek, Daphné ULg; Quevauvillers, Samuel ULg; Hebert, Jacques ULg et al

in Forêt Wallonne (2013), (124), 20-27

G-Cube, le logiciel de cubage et de constitution de lots de bois développé par Gembloux Agro-Bio Tech, fait peau neuve. Cette version 2.0 voit arriver des nouveautés parmi lesquelles : la géolocalisation ... [more ▼]

G-Cube, le logiciel de cubage et de constitution de lots de bois développé par Gembloux Agro-Bio Tech, fait peau neuve. Cette version 2.0 voit arriver des nouveautés parmi lesquelles : la géolocalisation des arbres cubés, le croisement des données avec des listes de prix personnelles ou standardisées, ou encore l’abandon de l’environnement Excel au profit d’une interface autonome. [less ▲]

Detailed reference viewed: 128 (29 ULg)
Full Text
See detailG-Cube version 2.3.1 - Logiciel de cubage et de lotissement d'arbres
Lejeune, Philippe ULg; Quevauvillers, Samuel ULg; Handerek, Daphné ULg

Software (2014)

This software is dedicated to the volume estimation of tree list using as set of volume equations adapted to belgian forest species. A specific module takes in charge the creation and the price ... [more ▼]

This software is dedicated to the volume estimation of tree list using as set of volume equations adapted to belgian forest species. A specific module takes in charge the creation and the price calculation of trees assortments [less ▲]

Detailed reference viewed: 1310 (103 ULg)
Full Text
Peer Reviewed
See detailg-Mode Oscillations in Hot B Subdwarf Stars
Charpinet, Stéphane; Van Grootel, Valérie ULg; Fontaine, Gilles et al

in Astronomical Society of the Pacific Conference Series (2013, December), 479

Observations from space with CoRoT and Kepler have revealed the rich pulsation spectra that characterize the long period hot B subdwarf (sdB) pulsators of the V1093 Her type. These evolved and compact ... [more ▼]

Observations from space with CoRoT and Kepler have revealed the rich pulsation spectra that characterize the long period hot B subdwarf (sdB) pulsators of the V1093 Her type. These evolved and compact helium core burning stars can develop low amplitude oscillations with periods in the range ∼ 1 − 4 hours that correspond to low degree, mid to high order g-modes. These modes have the property to penetrate deep inside the star, down to the boundary of the central convective core. They are consequently particularly interesting as asteroseismic probes. Here we briefly review how the g-modes behave in sdB stars, focusing in particular on the seismic signatures coming from the core itself. We also briefly summarize the first seismic results obtained from space data in pioneering efforts to exploit the potential of long period sdB pulsators as deep probes of the core structure. [less ▲]

Detailed reference viewed: 14 (0 ULg)
Full Text
See detailg-mode pulsations in slowly pulsating B stars
De Cat, P.; Daszynska-Daszkiewicz, J.; Briquet, Maryline ULg et al

in IAU Colloq. 193: Variable Stars in the Local Group (2004, May 01)

In this paper, we give an updated overview of the observed characteristics of g-mode pulsations in slowly pulsating B stars. These characteristics are based on the combined results of linear nonadiabatic ... [more ▼]

In this paper, we give an updated overview of the observed characteristics of g-mode pulsations in slowly pulsating B stars. These characteristics are based on the combined results of linear nonadiabatic analysis of the oscillations and a photometric and spectroscopic mode identification for a sample of 13 members. For 4 stars, at least one mode is stable in all the considered theoretical models. [less ▲]

Detailed reference viewed: 32 (2 ULg)
Full Text
Peer Reviewed
See detailG-mode trapping and period spacings in hot B subdwarf stars
Charpinet, Stéphane; Van Grootel, Valérie ULg; Brassard, Pierre et al

in Proceedings of the International Astronomical Union (2014, February), 301

Hot B subdwarfs (sdB) are hot and compact helium core burning stars of nearly half a solar mass that can develop pulsational instabilities driving acoustic and/or gravity modes. These evolved stars are ... [more ▼]

Hot B subdwarfs (sdB) are hot and compact helium core burning stars of nearly half a solar mass that can develop pulsational instabilities driving acoustic and/or gravity modes. These evolved stars are expected to be chemically stratified with an almost pure hydrogen envelope surrounding a helium mantle on top of a carbon/oxygen enriched core. However, the sdB stars pulsating in g-modes show regularities in their observed period distributions that, surprisingly (at first sight), are typical of the behavior of high order g-modes in chemically homogeneous (i.e., non-stratified) stars. This led to a claim that hot B subdwarfs could be much less chemically stratified than previously thought. Here, we reinvestigate trapping effects affecting g-modes in sdB stars. We show that standard stratified models of such stars can also produce nearly constant period spacings in the low frequency range similar to those found in g-mode spectra of sdB stars monitored with Kepler. [less ▲]

Detailed reference viewed: 11 (0 ULg)