Dissertations / Theses on the topic 'Ensembles'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Ensembles.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
Abbasian, Houman. "Inner Ensembles: Using Ensemble Methods in Learning Step." Thèse, Université d'Ottawa / University of Ottawa, 2014. http://hdl.handle.net/10393/31127.
Full textBenkaddour, Saïd. "Relation entre ensembles totalement flous et ensembles ordonnés." Lyon 1, 1986. http://www.theses.fr/1986LYO11714.
Full textBenkaddour, Saïd. "Relation entre les ensembles totalement flous et ensembles ordonnés." Grenoble 2 : ANRT, 1986. http://catalogue.bnf.fr/ark:/12148/cb37595915s.
Full textAndrle, Miroslav. "Ensembles modèles et analyse en ondelettes adaptées." Paris 7, 2002. http://www.theses.fr/2002PA077202.
Full textKhatib, Souad El. "Espaces métriques dans la théorie des ensembles flous." Lyon 1, 1986. http://www.theses.fr/1986LYO10060.
Full textDe, Jongh Albert. "Neural network ensembles." Thesis, Stellenbosch : Stellenbosch University, 2004. http://hdl.handle.net/10019.1/50035.
Full textENGLISH ABSTRACT: It is possible to improve on the accuracy of a single neural network by using an ensemble of diverse and accurate networks. This thesis explores diversity in ensembles and looks at the underlying theory and mechanisms employed to generate and combine ensemble members. Bagging and boosting are studied in detail and I explain their success in terms of well-known theoretical instruments. An empirical evaluation of their performance is conducted and I compare them to a single classifier and to each other in terms of accuracy and diversity.
AFRIKAANSE OPSOMMING: Dit is moontlik om op die akkuraatheid van 'n enkele neurale netwerk te verbeter deur 'n ensemble van diverse en akkurate netwerke te gebruik. Hierdie tesis ondersoek diversiteit in ensembles, asook die meganismes waardeur lede van 'n ensemble geskep en gekombineer kan word. Die algoritmes "bagging" en "boosting" word in diepte bestudeer en hulle sukses word aan die hand van bekende teoretiese instrumente verduidelik. Die prestasie van hierdie twee algoritmes word eksperimenteel gemeet en hulle akkuraatheid en diversiteit word met 'n enkele netwerk vergelyk.
Dromey, Christopher. "The Pierrot ensembles." Thesis, King's College London (University of London), 2010. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.682572.
Full textAl-Razgan, Muna Saleh. "Weighted clustering ensembles." Fairfax, VA : George Mason University, 2008. http://hdl.handle.net/1920/3212.
Full textVita: p. 134. Thesis director: Carlotta Domeniconi. Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Information Technology. Title from PDF t.p. (viewed Oct. 14, 2008). Includes bibliographical references (p. 128-133). Also issued in print.
Daigle, Elise. "Examining Music Ensemble Recruitment and Retention through Student Persistence into College Performing Ensembles." The Ohio State University, 2018. http://rave.ohiolink.edu/etdc/view?acc_num=osu1523523995130136.
Full textZajicek, Daniel J. "A rhetorical guide to Ebb." connect to online resource, 2006. http://www.unt.edu/etd/all/May2006/zajicek%5Fdaniel/index.htm.
Full textFor flute, oboe, clarinet, contrabassoon, horn , tuba, percussion (2 players), 2 violins, viola, cello, double bass, 1 laptop computer, and 2 or more loudspeakers. System requirements: Adobe Acrobat Reader. Duration: 10:00. Includes performance notes by the composer (p. 1-23). Includes bibliographical references (p. 23-24).
Shoemaker, Larry. "Ensembles for distributed data." [Tampa, Fla.] : University of South Florida, 2005. http://purl.fcla.edu/fcla/etd/SFE0001296.
Full textDubbs, Alexander. "Beta-ensembles with covariance." Thesis, Massachusetts Institute of Technology, 2014. http://hdl.handle.net/1721.1/90185.
Full text67
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 73-79).
This thesis presents analytic samplers for the [beta]-Wishart and [beta]-MANOVA ensembles with diagonal covariance. These generalize the [beta]-ensembles of Dumitriu-Edelman, Lippert, Killip-Nenciu, Forrester-Rains, and Edelman-Sutton, as well as the classical [beta] = 1, 2,4 ensembles of James, Li-Xue, and Constantine. Forrester discovered a sampler for the [beta]-Wishart ensemble around the same time, although our proof has key differences. We also derive the largest eigenvalue pdf for the [beta]-MANOVA case. In infinite-dimensional random matrix theory, we find the moments of the Wachter law, and the Jacobi parameters and free cumulants of the McKay and Wachter laws. We also present an algorithm that uses complex analysis to solve "The Moment Problem." It takes the first batch of moments of an analytic, compactly-supported distribution as input, and it outputs a fine discretization of that distribution.
by Alexander Dubbs.
Ph. D.
Roberts, Ekaterina Sergeevna. "Tailored random graph ensembles." Thesis, King's College London (University of London), 2014. https://kclpure.kcl.ac.uk/portal/en/theses/tailored-random-graph-ensembles(daefc925-24a3-4f7f-8c79-21b9136c636b).html.
Full textLlerena, Nils Ever Murrugarra. "Ensembles na classificação relacional." Universidade de São Paulo, 2011. http://www.teses.usp.br/teses/disponiveis/55/55134/tde-18102011-095113/.
Full textIn many fields, besides information about the objects or entities that compose them, there is also information about the relationships between objects. Some of these fields are, for example, co-authorship networks and Web pages. Therefore, it is natural to search for classification techniques that take into account this information. Among these techniques are the so-called graphbased classification, which seek to classify examples taking into account the relationships between them. This paper presents the development of methods to improve the performance of graph-based classifiers by using strategies of ensembles. An ensemble classifier considers a set of classifiers whose individual predictions are combined in some way. This combined classifier usually performs better than its individual classifiers. Three techniques have been developed: the first applied for originally propositional data transformed to relational format based on graphs and the second and the third applied for data originally in graph format. The first technique, inspired by the boosting algorithm originated the Adaptive Graph-Based K-Nearest Neighbor (A-KNN). The second technique, inspired by the bagging algorithm led to three approaches of Graph-Based Bagging (BG). Finally the third technique, inspired by the Cross- Validated Committees algorithm led to the Graph-Based Cross-Validated Committees (CVCG). The experiments were performed on 38 data sets, 22 datasets in propositional format and 16 in relational format. Evaluation was performed using the scheme of 10-fold stratified cross-validation and to determine statistical differences between the classifiers it was used the method proposed by Demsar (2006). Regarding the results, these three techniques improved or at least maintain the performance of the base classifiers. In conclusion, ensembles applied to graph-based classifiers have good results in the performance of them
Rosaz, Laurent. "Ensembles inevitables de mots." Paris 7, 1993. http://www.theses.fr/1993PA077093.
Full textPruniéras, Jean. "Ensembles [oméga] et applications." Paris 13, 2001. http://www.theses.fr/2001PA132024.
Full textThis thesis defines the set concept [omega] and proposes its application to problems which are not directly linked to each other
Swanzy, David Paul. "The wind ensemble and its music during the French Revolution (1789-1795) /." Ann Arbor : UMI, 2000. http://catalogue.bnf.fr/ark:/12148/cb370598520.
Full textMatsukevich, Dzmitry. "Quantum networking with atomic ensembles." Diss., Available online, Georgia Institute of Technology, 2006, 2006. http://etd.gatech.edu/theses/available/etd-07072006-173336/.
Full textKennedy, Brian, Committee Member ; Chapman, Michael, Committee Member ; Kuzmich, Alex, Committee Chair ; Raman, Chandra, Committee Member ; Voss, Paul, Committee Member.
Nunn, Joshua. "Quantum Memory in Atomic Ensembles." Thesis, University of Oxford, 2008. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.526095.
Full textHornik, Kurt. "A CLUE for CLUster ensembles." Institut für Statistik und Mathematik, WU Vienna University of Economics and Business, 2005. http://epub.wu.ac.at/1690/1/document.pdf.
Full textSeries: Research Report Series / Department of Statistics and Mathematics
Fisher, Charles. "Statistical Characterization of Protein Ensembles." Thesis, Harvard University, 2012. http://dissertations.umi.com/gsas.harvard:10223.
Full textSiegert, Stefan. "Rank statistics of forecast ensembles." Doctoral thesis, Saechsische Landesbibliothek- Staats- und Universitaetsbibliothek Dresden, 2013. http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa-102152.
Full textDumitriu, Ioana 1976. "Eigenvalue statistics for beta-ensembles." Thesis, Massachusetts Institute of Technology, 2003. http://hdl.handle.net/1721.1/29347.
Full textIncludes bibliographical references (p. 155-163) and index.
Random matrix theory is a maturing discipline with decades of research in multiple fields now beginning to converge. Experience has shown that many exact formulas are available for certain matrices with real, complex, or quaternion entries. In random matrix jargon, these are the cases β = 1, 2 and 4 respectively. This thesis explores the general P > 0 case mathematically and with symbolic software. We focus on generalizations of the Hermite distributions originating in physics (the "Gaussian" ensembles) and the Laguerre distributions of statistics (the "Wishart" matrices). One of our main contributions is the construction of tridiagonal matrix models for the general (β > 0) 0 β-Hermite and (β > 0, a > β(m - 1)/2) β-Laguerre ensembles of parameter a and size m, and investigate applications of these new ensembles, particularly in the areas of eigenvalue statistics. The new models are symmetric tridiagonal, and with entries from real distributions, regardless of the value of β. The entry distributions are either normal or X, so "classical", and the independence pattern is maximal, in the sense that the only constraints arise from the symmetric/semi-definite condition. The β-ensemble distributions have been studied for the particular 1, 2, 4 values of p as joint eigenvalue densities for full random matrix ensembles (Gaussian, or Hermite, and Wishart, or Laguerre) with real, complex, and quaternion entries (for references, see [66] and [70]). In addition, general -ensembles were considered and studied as theoretical distributions ([8, 51, 50, 55, 56]), with applications in lattice gas theory and statistical mechanics (the parameter being interpreted as an arbitrary inverse temperature of a Coulomb gas with logarithmic potential).
(cont.) Certain eigenvalue statistics over these general β-ensembles, namely those expressible in terms of integrals of symmetric polynomials with corresponding Hermite or Laguerre weights, can be computed in terms of multivariate orthogonal polynomials (Hermite or Laguerre). We have written a Maple Library (MOPs: Multivariate Orthogonal Polynomials symbolically) which implements some new and some known algorithms for computing the Jack, Hermite, Laguerre, and Jacobi multivariate polynomials for arbitrary. This library can be used as a tool for conjecture-formulation and testing, for statistical computations, or simply for getting acquainted with the mathematical concepts. Some of the figures in this thesis have been obtained using MOPs. Using the new β-ensemble models, we have been able to provide a unified perspective of the previously isolated 1, 2, and 4 cases, and prove generalizations for some of the known eigenvalue statistics to arbitrary β. We have rediscovered (in the Hermite case) a strong version of the Wigner Law (semi-circle), and proved (in the Laguerre case) a strong version of the similar law (generalized quarter-circle). We have obtained first-order perturbation theory for the P large case, and we have reason to believe that the tridiagonal models in the large n (ensemble size) limit will also provide a link between the largest eigenvalue distributions for both Hermite and Laguerre for arbitrary P (for β = 1, 2, this link was proved to exist by Johannson [52] and Johnstone [53]) ...
by Ioana Dumitriu.
Ph.D.
Moupila-Mapépé, Aurélien. "Hyperbolicité et ensembles invariants isolés." Mémoire, Université de Sherbrooke, 2001. http://savoirs.usherbrooke.ca/handle/11143/4529.
Full textBrown, Gavin. "Diversity in neural network ensembles." Thesis, University of Birmingham, 2004. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.410855.
Full textFernandes, Everlandio Rebouças Queiroz. "Evolutionary ensembles for imbalanced learning." Universidade de São Paulo, 2018. http://www.teses.usp.br/teses/disponiveis/55/55134/tde-30102018-101256/.
Full textEm muitos problemas reais de classificação, o conjunto de dados usado para a indução do modelo é significativamente desbalanceado. Isso ocorre quando a quantidade de exemplos de algumas classes é muito inferior às das outras classes. Conjuntos de dados desbalanceados podem comprometer o desempenho da maioria dos algoritmos clássicos de classificação. Os modelos de classificação induzidos por tais conjuntos de dados geralmente apresentam um forte viés para as classes majoritárias, tendendo classificar novas instâncias como pertencentes a essas classes. Uma estratégia comumente adotada para lidar com esse problema, é treinar o classificador sobre uma amostra balanceada do conjunto de dados original. Entretanto, esse procedimento pode descartar exemplos que poderiam ser importantes para uma melhor discriminação das classes, diminuindo a eficiência do classificador. Por outro lado, nos últimos anos, vários estudos têm mostrado que em diferentes cenários a estratégia de combinar vários classificadores em estruturas conhecidas como comitês tem se mostrado bastante eficaz. Tal estratégia tem levado a uma acurácia preditiva estável e principalmente a apresentar maior habilidade de generalização que os classificadores que compõe o comitê. Esse poder de generalização dos comitês de classificadores tem sido foco de pesquisas no campo de aprendizado desbalanceado, com o objetivo de diminuir o viés em direção as classes majoritárias, apesar da complexidade que envolve gerar comitês de classificadores eficientes. Meta-heurísticas de otimização, como os algoritmos evolutivos, têm muitas aplicações para o aprendizado de comitês, apesar de serem pouco usadas para este fim. Por exemplo, algoritmos evolutivos mantêm um conjunto de soluções possíveis e diversificam essas soluções, o que auxilia na fuga dos ótimos locais. Nesse contexto, esta tese investiga e desenvolve abordagens para lidar com conjuntos de dados desbalanceados, utilizando comitês de classificadores induzidos a partir de amostras do conjunto de dados original por meio de metaheurísticas. Mais especificamente, são propostas três soluções baseadas em aprendizado evolucionário de comitês e uma quarta proposta que utiliza um mecanismo de poda baseado em ranking de dominância, conceito comum em algoritmos evolutivos multiobjetivos. Experimentos realizados mostraram o potencial das soluções desenvolvidas.
Kirchner, Ulrich. "Approaches to ensembles of universes." Doctoral thesis, University of Cape Town, 2003. http://hdl.handle.net/11427/4898.
Full textThis thesis consists of three parts - each of them focusing on different aspects relating to ensembles of universes and causally disconnected regions. In the first part I investigate possible measures over the space of FLRW models. In the second part I examine the behaviour of such transition regions for spherically symmetric space-times. In the last part of this thesis I discuss philosophical, physical, and probabilistic issues related to the concept of a multiverse - an ensemble of universes. The difference between ensembles of really existing universes and ensembles of possible universes is emphasized.
Berglund, Filip. "Asymptotics of beta-Hermite Ensembles." Thesis, Linköpings universitet, Matematisk statistik, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-171096.
Full textI denna kandidatuppsats presenterar vi resultat om några olika egenvärdens-statistikor från beta-Hermite ensemblerna, först i de klassiska fallen då beta = 1, 2, 4, det vill säga den gaussiska ortogonala ensemblen (bestående av reella symmetriska matriser), den gaussiska unitära ensemblen (bestående av komplexa hermitiska matriser) och den gaussiska symplektiska ensemblen (bestående av kvaternioniska själv-duala matriser). Vi tittar även på de mindre undersökta generella beta-Hermite ensemblerna (bestående av reella symmetriska tridiagonala matriser). Specifikt tittar vi på den empiriska fördelningsfunktionen och två olika normeringar av det största egenvärdet. De resultat vi presenterar för dessa statistikor är den empiriska fördelningsfunktionens konvergens mot halvcirkel-fördelningen, det normerade största egenvärdets konvergens mot Tracy-Widom fördelningen, och, med en annan normering, största egenvärdets konvergens mot 1. Vi illustrerar även dessa resultat med hjälp av simuleringar. För den gaussiska unitära ensemblen presenterar vi ett uttryck för dess nivåtäthet. För att underlätta förståelsen av den gaussiska symplektiska ensemblen presenterar vi egenskaper hos egenvärdena av kvaternioniska matriser. Slutligen bevisar vi en sats om symmetrin hos ordningsstatistikan av egenvärdena av beta-Hermite ensemblerna.
Baffoe, Nana Ama Appiaa. "Diagnostic Tools for Forecast Ensembles." Case Western Reserve University School of Graduate Studies / OhioLINK, 2018. http://rave.ohiolink.edu/etdc/view?acc_num=case1522964882574611.
Full textRahman, Md Ahsanur. "Unstable Communities in Network Ensembles." Diss., Virginia Tech, 2016. http://hdl.handle.net/10919/78290.
Full textPh. D.
Grumbach, Stéphane. "Ensembles et fonctions en Datalog." Paris 11, 1990. http://www.theses.fr/1990PA112003.
Full textTardivel, Valérie. "Ensembles coanalytiques en analyse harmonique." Paris 6, 1986. http://www.theses.fr/1986PA066096.
Full textLachièze-Rey, Raphaël. "Ensembles convexes et mosaïques aléatoires." Thesis, Lille 1, 2010. http://www.theses.fr/2010LIL10084/document.
Full textThis thesis is concerned with various problems of stochastic geometry, more particularly inspired from convex geometry. The first chapter is devoted to the study of extreme points of the graph of some Markov processes. Then we are interestd in the convex rearrangement of centered Gaussian fields. In the third chapter we study the ergodicity of STIT tessellations, a particular class of random mosaics in R^d. Finally, we the study the peeling of a Point process, with the generalisation of the notion of convex hull to that of L-enveloppe, with L an arbitrary class of closed sets
Cadena, Carlos. "Evolutionary feature creation for ensembles." Thèse, Montréal : École de technologie supérieure, 2008. http://proquest.umi.com/pqdweb?did=1472144241&sid=4&Fmt=2&clientId=46962&RQT=309&VName=PQD.
Full text"by Cadena, Carlos" -- p. de. t. "Thesis presented to École de technologie supérieure in partial fulfillment of the requirements for the degree of master of engineering." CAQMUQET CaQMUQET Bibliogr. : f.[223]-229. Également disponible en version électronique. CaQMUQET
Moupila-Mapépé, Aurélien. "Hyperbolicité et ensembles invariants isolés." Sherbrooke : Université de Sherbrooke, 2001.
Find full textTardivel, Valérie. "Ensembles coanalytiques en analyse harmonique." Grenoble 2 : ANRT, 1986. http://catalogue.bnf.fr/ark:/12148/cb37601475m.
Full textLupu, Titus. "Ensembles poissoniens de boucles markoviennes." Thesis, Paris 11, 2015. http://www.theses.fr/2015PA112066/document.
Full textIn this thesis I study an infinite measure on loops naturally associated to a wide range of Markovian processes and the Poisson point processes of intensity proportional to this measure (intensity parameter alpha>0). This Poissson point processes are called Poisson ensembles of Markov loops or loop soups. The measure on loops is covariant with some transformation on Markovian processes, for instance the change of time. In the setting of Brownian loop soups inside a proper open simply connected domain of C it was shown that the outer boundaries of outermost clusters of loops are, for alpha1/2, Conformal Loop Ensembles CLE(kappa), kappa in (8/3,4]. Besides, it was shown for a wide range of symmetric Markovian processes that for alpha=1/2 the occupation field of a loop soup (the sum of times spent by loops over points) is the square of the Gaussian free field. First I studied the loop soups associated to one-dimensional diffusions, and particularly the occupation field and its zeroes that delimit in this case the clusters of loops. Then I studied the loop soups on discrete graphs and metric graphs (edges replaced by continuous lines). On a metric graph on one hand the loops have a non-trivial geometry and on the other hand one has the same property as in the setting of one-dimensional diffusions that the zeroes of the occupation field delimit the clusters of loops. By combing metric graphs and the isomorphism with the Gaussian free field I have shown that alpha=1/2 is the critical parameter for random walk loop soup percolation on the discrete half-plane Z*N (existence or not of an infinite cluster of loops) and that for alpha<= 1/2 the scaling limit of outer boundaries of outermost clusters on Z*N is a CLE(kappa) on the continuum half plane
Zagorski, Marcus. "And drown the wakeful anguish of the soul." Thesis, National Library of Canada = Bibliothèque nationale du Canada, 1998. http://www.collectionscanada.ca/obj/s4/f2/dsk1/tape11/PQDD_0007/MQ43989.pdf.
Full textLecomte, Dominique. "Classes de Wadge potentielles et théorèmes d'uniformisation partielle." Paris 6, 1993. http://www.theses.fr/1993PA066591.
Full textColangelo, Giorgio. "Quantum measurements with cold atomic ensembles." Doctoral thesis, Universitat Politècnica de Catalunya, 2016. http://hdl.handle.net/10803/404449.
Full textEsta tesis describe medidas cuánticas en una nube de átomos fríos de rubidio-87. Extendemos el formalismo de matriz de covariancia a sistemas de espin-1, incluyendo decoherencias, pérdidas debidas a la luz de prueba y fluctuaciones del número de átomos. Mostramos que el modelo puede reproducir los resultados experimentales de medias y variancias de medidas de rotación de Faraday de una señal de decaimiento libre de inducción de un estado coherente de spin rotando en un campo magnético. Derivamos procedimientos de linearizacion para medidas de Faraday con ángulos de rotación grandes y desarrollamos un fotodetector rápido que permite la detección en rango dinámico alto. En esta tesis estudiamos también el problema de interacción luz-átomos en el caso de acoplamiento inhomogéneo y calibramos experimentalmente el nivel de ruido cuántico. Demostramos que dos observables colectivas de espín que describen la nube atómica que no conmutan se pueden conocer simultáneamente con precisión más allá de los límites clásicos, produciendo un estado cuántico planar comprimido. Estudiamos teóricamente sus ventajas metrológicas y encontramos condiciones óptimas para su realización. Así mismo, utilizando medidas cuánticas no desctructivas (QND) en un estado atómico coherente rotando en un campo magnético ortogonal, monitorizamos las componentes radial y angular del spin colectivo de la nube de átomos 7.0 dB por debajo de la estadística Poissoniana y 2.9 dB por debajo del nivél del ruido de proyección. El último resultado de esta tesis es la investigación del contra-efecto de la medida como un auténtico efecto cuántico a través de la violación de las desigualdades de Leggett-Garg. El uso de las medidas QND, que no perturban la observable medida, permite considerar las violaciones como en verdadero efecto cuántico en lugar de otros posibles efectos clásicos causados por la medida. Por otro lado el uso de estados Gaussianos usado en los cálculos de matriz de covariancia puede ser aplicado también a sistemas verdaderamente macroscópicos.
Montbrió, i. Fairen Ernest. "Synchronization in ensembles of nonisochronous oscillators." [S.l. : s.n.], 2004. http://pub.ub.uni-potsdam.de/2004/0049/mtbrio.pdf.
Full textZell, Thierry. "Etude quantitative des ensembles semi-pfaffiens." Phd thesis, Université Rennes 1, 2003. http://tel.archives-ouvertes.fr/tel-00008488.
Full textZhou, Da Sheng. "Eigenvalues statistics for restricted trace ensembles." Thesis, University of Macau, 2010. http://umaclib3.umac.mo/record=b2182958.
Full textMontbrió, i. Fairen Ernest. "Synchronization in ensembles of nonisochronous oscillators." Phd thesis, Universität Potsdam, 2004. http://opus.kobv.de/ubp/volltexte/2005/158/.
Full textThis thesis analyses synchronization phenomena occurring in large ensembles of interacting oscillatory units. In particular, the effects of nonisochronicity (frequency dependence on the oscillator's amplitude) on the macroscopic transition to synchronization are studied in detail. The new phenomena found (Anomalous Synchronization) are investigated in populations of oscillators as well as between oscillator's ensembles.
Pisetta, Vincent. "New Insights into Decision Trees Ensembles." Thesis, Lyon 2, 2012. http://www.theses.fr/2012LYO20018/document.
Full textDecision trees ensembles are among the most popular tools in machine learning. Nevertheless, their theoretical properties as well as their empirical performances are subject to strong investigation up to date. In this thesis, we propose to shed light on these methods. More precisely, after having described the current theoretical aspects of three main ensemble schemes (chapter 1), we give an analysis supporting the existence of common reasons to the success of these three principles (chapter 2). This last takes into account the two first moments of the margin as an essential ingredient to obtain strong learning abilities. Starting from this rejoinder, we propose a new ensemble algorithm called OSS (Oriented Sub-Sampling) whose steps are in perfect accordance with the point of view we introduce. The empirical performances of OSS are superior to the ones of currently popular algorithms such as Random Forests and AdaBoost. In a third chapter (chapter 3), we analyze Random Forests adopting a “kernel” point of view. This last allows us to understand and observe the underlying regularization mechanism of these kinds of methods. Adopting the kernel point of view also enables us to improve the predictive performance of Random Forests using popular post-processing techniques such as SVM and multiple kernel learning. In conjunction with random Forests, they show greatly improved performances and are able to realize a pruning of the ensemble by conserving only a small fraction of the initial base learners
Ardeshir, G. "Decision tree simplification for classifier ensembles." Thesis, University of Surrey, 2002. http://epubs.surrey.ac.uk/843022/.
Full textNagel, Uwe [Verfasser]. "Analysis of Network Ensembles / Uwe Nagel." Konstanz : Bibliothek der Universität Konstanz, 2013. http://d-nb.info/1030479054/34.
Full textAhmad, Amir. "Data Transformation for Decision Tree Ensembles." Thesis, University of Manchester, 2009. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.508528.
Full textPlumpton, Catrin Oliver. "Classifier ensembles for streaming fMRI data." Thesis, Bangor University, 2011. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.540419.
Full textBilgic̦, Berkin. "Fast human detection with cascaded ensembles." Thesis, Massachusetts Institute of Technology, 2010. http://hdl.handle.net/1721.1/57684.
Full textCataloged from PDF version of thesis.
Includes bibliographical references (p. 75-78).
Detecting people in images is a challenging task because of the variability in clothing and illumination conditions, and the wide range of poses that people can adopt. To discriminate the human shape clearly, Dalal and Triggs [1] proposed a gradient based, robust feature set that yielded excellent detection results. This method computes locally normalized gradient orientation histograms over blocks of size 16x16 pixels representing a detection window. The block histograms within the window are then concatenated. The resulting feature vector is powerful enough to detect people with 88% detection rate at 10 -4 false positives per window (FPPW) using a linear SVM. The detection window slides over the image in all possible image scales; hence this is computationally expensive, being able to run at 1 FPS for a 320x240 image on a typical CPU with a sparse scanning methodology. Due to its simplicity and high descriptive power, several authors worked on the Dalal-Triggs algorithm to make it feasible for real time detection. One such approach is to implement this method on a Graphics Processing Unit (GPU), exploiting the parallelisms in the algorithm. Another way is to formulate the detector as an attentional cascade, so as to allow early rejections to decrease the detection time. Zhu et al. [2] demonstrated that it is possible to obtain a 30x speed up over the original algorithm with this methodology.
(cont.) In this thesis, we combine the two proposed methods and investigate the feasibility of a fast person localization framework that integrates the cascade-of-rejectors approach with the Histograms of Oriented Gradients (HoG) features on a data parallel architecture. The salient features of people are captured by HoG blocks of variable sizes and locations which are chosen by the AdaBoost algorithm from a large set of possible blocks. We use the integral image representation for histogram computation and a rejection cascade in a sliding-windows manner, both of which can be implemented in a data parallel fashion. Utilizing the NVIDIA CUDA framework to realize this method on a Graphics Processing Unit (GPU), we report a speed up by a factor of 13 over our CPU implementation. For a 1280x960 image our parallel technique attains a processing speed of 2.5 to 8 frames per second depending on the image scanning density, with a detection quality comparable to the original HoG algorithm.
by Berkin Bilgic̦.
S.M.