To see the other types of publications on this topic, follow the link: Gibbir.
Dissertations / Theses on the topic 'Gibbir'
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 'Gibbir.'
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.
1
Wako, Adi Liban, University of Western Sydney, College of Social and Health Sciences, and School of Applied Social and Human Sciences. "Ideology as commodity : industry of a theocracy and production of famines in Ethiopia." THESIS_CSHS_ASH_WakoAdi-L.xml, 2003. http://handle.uws.edu.au:8081/1959.7/452.
Full textAbstract:
This work introduces an alternative paradigm that claims that the primary industry of imperial Ethiopia has been (and still is) one that has evolved around the Abyssinian national mission vis-a-vis the populations it rules from a point of conquest. Abyssinia, like Catholic Spain in the Americas, carried out a series of 'civilising' missions (Christianising), that has spanned centuries to modern times. Around a theocratic mission evolved a service type industry, the author calls theo-industry. On that basis, the work demonstrates how well-known categories of 'land-tenure', namely, the gult/goolt, the gultenya/gooltenya, the rist, the ristenya, the gabbar and related others are categories of a fiscal system of theo-industry, not of an agrarian system or agrarian industry. It is argued how these rather complex categories belong in the realm of wages and pensions of a service-type industry, not in those of agriculture. By failing to establish the functional link between agriculture and the national mission of the rulers, the scholars of Ethiopian studies have so far been unable to identify this 'elusive' but all-pervasive primary industry of Ethiopia. That in turn, the author argues, has had a rub-off effect in hindering a clear and comprehensive understanding of issues such as poverty and famine. The central topic of this work is the 'identification' of this 'elusive' industry. The study of its evolution, set in historical grounds, of its dynamics and the intricate maze of multi-natured relations is attempted. On this basis, the option of creating an independent (from theo-industry), and more importantly, renewable agricultural industry is proposed as the key to tackling chronic levels of poverty and famine in EthiopiaDoctor of Philosophy (PhD)
2
Johnston, Ian Andrew. "Gibber." Thesis, University of Dundee, 1988. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.252137.
Full text
3
Wako, Adi Liban. "Ideology as commodity : industry of a theocracy and production of famines in Ethiopia." Thesis, View thesis, 2003. http://handle.uws.edu.au:8081/1959.7/452.
Full textAbstract:
This work introduces an alternative paradigm that claims that the primary industry of imperial Ethiopia has been (and still is) one that has evolved around the Abyssinian national mission vis-a-vis the populations it rules from a point of conquest. Abyssinia, like Catholic Spain in the Americas, carried out a series of 'civilising' missions (Christianising), that has spanned centuries to modern times. Around a theocratic mission evolved a service type industry, the author calls theo-industry. On that basis, the work demonstrates how well-known categories of 'land-tenure', namely, the gult/goolt, the gultenya/gooltenya, the rist, the ristenya, the gabbar and related others are categories of a fiscal system of theo-industry, not of an agrarian system or agrarian industry. It is argued how these rather complex categories belong in the realm of wages and pensions of a service-type industry, not in those of agriculture. By failing to establish the functional link between agriculture and the national mission of the rulers, the scholars of Ethiopian studies have so far been unable to identify this 'elusive' but all-pervasive primary industry of Ethiopia. That in turn, the author argues, has had a rub-off effect in hindering a clear and comprehensive understanding of issues such as poverty and famine. The central topic of this work is the 'identification' of this 'elusive' industry. The study of its evolution, set in historical grounds, of its dynamics and the intricate maze of multi-natured relations is attempted. On this basis, the option of creating an independent (from theo-industry), and more importantly, renewable agricultural industry is proposed as the key to tackling chronic levels of poverty and famine in Ethiopia
4
Eskelinen, Mareija. "Herded Gibbs and discretized herded Gibbs sampling." Thesis, University of British Columbia, 2013. http://hdl.handle.net/2429/44896.
Full textAbstract:
Statistical inference is at the heart of the probabilistic programming approach to artificial intelligence; as far as statistical inference is concerned, the Gibbs sampler is one of the most widely applicable and popular random sampling techniques. On the flip side, deterministic sampling is still nascent and has yet to be widely accepted into the field of Bayesian inference.
In this thesis, we advance the domain of deterministic sampling by discussing and evaluating two deterministic sampling techniques inspired by the Gibbs sampler. The first technique is the recently-introduced herded Gibbs sampler and the second is the novel discretized herded Gibbs sampler. Herded Gibbs sampling is an entirely deterministic algorithm with an impressive O(1/T) convergence rate for both models with independent variables and fully connected probabilistic graphical models. However, the convergence for herded Gibbs on sparsely connected probabilistic graphical models is still an open problem.
Additionally, the computational complexity of herded Gibbs increases exponentially with the connectivity of the probabilistic graphical model, making it better suited for partially connected graphical models. We empirically demonstrate that herded Gibbs outperforms Gibbs on partially connected probabilistic graphical models for the tasks of image denoising with Markov Random Fields and named entity recognition with Conditional Random Fields. Furthermore, we develop discretized herded Gibbs as a variant of herded Gibbs that attempts to reduce the spatial complexity by approximating herded Gibbs. Though discretized herded Gibbs achieves linear storage requirements and is capable of fast initial convergence rates, it can only approximate the target up to a finite accuracy. This thesis highlights the power of one recent deterministic sampling algorithm and
shows that deterministic sampling with good space and time complexity remains an elusive goal.
5
Morisi, Silvia. "L'effetto Gibbs." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2015. http://amslaurea.unibo.it/9722/.
Full textAbstract:
In questa tesi studiamo l'effetto Gibbs. Tale fenomeno si manifesta tramite la presenza di sovra-oscillazioni nei polinomi di Fourier di funzioni che presentano discontinuità di prima specie. La differenza tra il massimo ed il minimo del polinomio di Fourier di tali funzioni, in prossimità di un punto di discontinuità della funzione, è strettamente maggiore del salto della funzione in quel punto, anche per n che tende all'infinito. Per attenuare le sovra-oscillazioni delle somme parziali di Fourier si utilizzano le serie di Fejer e si vede come effettivamente il fenomeno di Gibbs scompaia.
6
Console, Sarah. "L'effetto Gibbs." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2018. http://amslaurea.unibo.it/15924/.
Full textAbstract:
In questa trattazione viene definito l'effetto Gibbs: un fenomeno di sovraoscillazione che si presenta nei polinomi di Fourier associati a funzioni periodiche, regolari a tratti con punti di discontinuità di prima specie. Si dimostra inoltre come eliminare l'effetto Gibbs utilizzando i polinomi di Fejér invece dei polinomi di Fourier.
7
Scarpellini, Damiano. "L'effetto Gibbs." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2020. http://amslaurea.unibo.it/22153/.
Full textAbstract:
L'effetto Gibbs è un fenomeno oscillatorio che riguarda i polinomi di Fourier. In particolare, ogni volta che si approssima una funzione periodica e regolare a tratti con il suo polinomio di Fourier associato, si può notare che, in prossimità dei punti di discontinuità di tipo salto di tale funzione, il polinomio presenta delle sovraelongazioni che non si attenuano all'aumentare del grado.
Lo studio di questo fenomeno è stato diviso in tre parti.
Nel primo capitolo vengono studiati i polinomi e le serie di Fourier e vengono forniti i principali risultati di convergenza. Il secondo capitolo introduce il fenomeno di Gibbs attraverso due esempi particolari e si conclude con l'analisi del caso generale per funzioni periodiche e regolari a tratti. Infine, nel terzo e ultimo capitolo, vengono studiate le somme di Fejér e si dimostra come queste siano una soluzione all'effetto Gibbs.
8
Mangianti, Marco. "L'effetto Gibbs." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2014. http://amslaurea.unibo.it/7904/.
Full textAbstract:
Nella tesi ho trattato l'effetto Gibbs,ovvero la presenza di forti oscillazioni nei polinomi di Fourier di funzioni con discontinuità di prima specie. Infine ho introdotto le somme di Fejér osservando come con questi polinomi trigonometrici si possa eliminare l'effetto Gibbs.
9
Roberts, Charlotte Olivia. "Edward Gibbon : history and identity." Thesis, University of Cambridge, 2012. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.610425.
Full text
10
Fang, Jing. "Herded Gibbs sampling." Thesis, University of British Columbia, 2012. http://hdl.handle.net/2429/43188.
Full textAbstract:
The Gibbs sampler is one of the most popular algorithms for inference in statistical models. In this thesis, we introduce a herding variant of this algorithm that is entirely deterministic. We demonstrate, with simple examples, that herded Gibbs exhibits better convergence behavior for approximating the marginal distributions than Gibbs sampling. In particular, image denoising exemplifies the effectiveness of herded Gibbs as an inference technique for Markov Random Fields (MRFs). Also, we adopt herded Gibbs as the inference engine for Conditional Random Fields (CRFs) in Named Entity Recognition (NER) and show that it is competitive with the state of the art. The conclusion is that herded Gibbs, for graphical models with nodes of low degree, is very close to Gibbs sampling in terms of the complexity of the code and computation, but that it converges much faster.
11
Salame, Younis Jarir. "Il fenomeno di Gibbs." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2016. http://amslaurea.unibo.it/12388/.
Full textAbstract:
L'argomento di questa tesi è il fenomeno di Gibbs, un fenomeno che riguarda dei particolari polinomi trigonometrici, i polinomi di Fourier: questi ultimi sono molto utilizzati per approssimare funzioni periodiche in quanto tra tutti i polinomi trigonometrici sono quelli che minimizzano l'errore in norma quadratica. Il fenomeno di Gibbs si può descrivere in questo modo: siano date una funzione f periodica che presenta punti di discontinuità di prima specie e la sua approssimazione tramite serie di Fourier; quando si tronca questa serie si avranno delle forti oscillazioni in prossimità dei punti di discontinuità della funzione. Aumentando il numero di componenti della serie troncata le oscillazioni si avvicinano sempre di più ai relativi punti di discontinuità ma il picco rimane lo stesso. La tesi è stata suddivisa in tre capitoli. Nel primo si inizia definendo i polinomi di Fourier e mostrando alcune loro proprietà, dopodiché si studia la convergenza della serie di Fourier. Nel secondo si descrive il fenomeno di Gibbs: si comincia mostrando graficamente cosa accade in tre esempi distinti di funzioni periodiche in prossimità dei loro punti di discontinuità; si analizza quindi quantitativamente uno di questi esempi e si conclude fornendo un risultato generale. Infine nel terzo, dopo aver definito le somme di Fejér e un nuovo tipo di convergenza, quella secondo Cesaro si mostra come correggere questo fenomeno utilizzando le somme di Fejér al posto dei polinomi di Fourier.
12
Monaldi, Alessio. "Il fenomeno di Gibbs." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2018. http://amslaurea.unibo.it/16863/.
Full textAbstract:
Studio della serie di Fourier e risultati di convergenza. Analisi matematica del fenomeno di Gibbs partendo da alcuni esempi, come quello dell'onda quadra. Analisi quantitativa del fenomeno, sviluppo grafico delle approssimazioni prese in esame attraverso algoritmi implementati in Matlab. Studio delle somme di Fejér e risoluzione del fenomeno di Gibbs.
13
Farina, Lorenzo. "Il fenomeno di Gibbs." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2019. http://amslaurea.unibo.it/19513/.
Full textAbstract:
Questa trattazione ha lo scopo di fornire una spiegazione matematica del fenomeno di Gibbs, il quale si verifica quando sono presenti forti oscillazioni nei polinomi di Fourier di una funzione con discontinuità di prima specie. Osserveremo che queste peculiarità, che si trovano vicino ai punti di discontinuità della funzione, non diminuiscono aumentando il grado del polinomio, tanto che la serie sembra non convergere alla funzione sviluppata. Vedremo che un modo per far scomparire questo fenomeno è quello di utilizzare un altro tipo di polinomi trigonometrici, cioè quelli di Fejér, al posto di quelli di Fourier. Di solito però si preferisce utilizzare comunque il polinomio di Fourier per la rappresentazione della funzione, anche se può portare a queste anomalie, perché è il polinomio trigonometrico che meglio approssima la funzione in norma quadratica. Diremo che lo studio di questo fenomeno è molto attuale, perché in tutti i processi di approssimazione, quali per esempio la ricostruzione o il filtraggio di immagini e segnali, bisogna tener controllate queste particolari oscillazioni descritte sopra, cercando di eliminarne il più possibile.
14
Poluzzi, Diletta. "Il fenomeno di Gibbs." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2021. http://amslaurea.unibo.it/23171/.
Full textAbstract:
In questa trattazione è presente una descrizione dettagliata del fenomeno di Gibbs , che si presenta quando vi sono delle forti oscillazioni dei polinomi di Fourier, con i quali una funzione regolare a tratti e periodica viene approssimata, attorno alle discontinuità di prima specie. Questo fenomeno risulta essere ancora estremamente attuale; infatti nei processi di approssimazione, riguardanti, per esempio, la ricostruzione o il filtraggio di immagini o segnali, è fondamentale riuscire a controllare le oscillazioni dovute a tale fenomeno. Ho fornito nozioni preliminari per descrivere in modo dettagliato il fenomeno preso in esame come la definizione di serie di Fourier ed i relativi criteri di convergenza (puntale, uniforme e in media quadratica). Infine ho definito le serie di Fejér e mostrato che, approssimando le funzioni con queste, il fenomeno di Gibbs non avviene.
15
Chimisov, Cyril. "Adapting the Gibbs sampler." Thesis, University of Warwick, 2018. http://wrap.warwick.ac.uk/108829/.
Full textAbstract:
In the present thesis, we close a methodological gap of optimising the basic Markov Chain Monte Carlo algorithms. Similarly to the straightforward and computationally efficient optimisation criteria for the Metropolis algorithm acceptance rate (and, equivalently, proposal scale), we develop criteria for optimising the selection probabilities of the Random Scan Gibbs Sampler. We develop a general purpose Adaptive Random Scan Gibbs Sampler, that adapts the selection probabilities, gradually, as further information is accrued by the sampler. We argue that Adaptive Random Scan Gibbs Samplers can be routinely implemented and substantial computational gains will be observed across many typical Gibbs sampling problems. Additionally, motivated to develop theory to analyse convergence properties of the Adaptive Gibbs Sampler, we introduce a class of Adapted Increasingly Rarely Markov Chain Monte Carlo (AirMCMC) algorithms, where the underlying Markov kernel is allowed to be changed based on the whole available chain output, but only at specific time points separated by an increasing number of iterations. The main motivation is the ease of analysis of such algorithms. Under regularity assumptions, we prove the Mean Square Error convergence, Weak and Strong Laws of Large Numbers, and the Central Limit Theorem and discuss how our approach extends the existing results. We argue that many of the known Adaptive MCMC algorithms may be transformed into the corresponding Air versions and provide an empirical evidence that performance of the Air version remains virtually the same.
16
Guidi, Chiara. "Il fenomeno di Gibbs." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2013. http://amslaurea.unibo.it/5668/.
Full textAbstract:
Questa trattazione si propone di fornire una spiegazione del fenomeno di Gibbs in termini matematici. Con l'espressione fenomeno di Gibbs intendiamo la presenza di forti oscillazioni nei polinomi di Fourier di una funzione con discontinuità di prima specie. Si osserva che queste anomalie, presenti vicino ai punti di discontinuità, non sembrano diminuire aumentando il grado del polinomio, al punto che la serie pare non convergere alla funzione sviluppata.
Osserveremo che utilizzando un altro tipo di polinomi trigonometrici, quelli di Fejér in luogo di quelli di Fourier, scomparirà il fenomeno di Gibbs. Nonostante ciò, spesso si preferisce rappresentare una funzione utilizzando il suo polinomio di Fourier poiché questo è il polinomio trigonometrico che meglio approssima la funzione in norma quadratica.
17
Meyer, Michael. "Gibbon, Mill und Ruskin : Autobiographie und Intertextualität /." Heidelberg : C. Winter, 1998. http://catalogue.bnf.fr/ark:/12148/cb38996938h.
Full text
18
Karrat, Mohamed. "Transformations de mesures de Gibbs." Lille 1, 2000. http://www.theses.fr/2000LIL10197.
Full textAbstract:
Nous considerons deux bases mesurables (, f, i, f) et ( *, f *, i *, f *) et une transformation (, ) f/f *-admissible. Nous montrons que si p est une mesure de gibbs associee a une f-specification = ( v) v e i satisfaisant la condition de dynkin (i. E. Les images des noyaux v, par , sont constantes sur les classes d'equivalence de ), alors sa mesure image p par est une mesure de gibbs associee a une f *-specification * obtenue a partir de par une transformation de renormalisation r ,. En particulier, lorsque est gibbsienne (i. E. Definie a partir d'un f-hamiltonien h et d'une f-specification de reference o), nous donnons des conditions suffisantes, sur h et o, qui garantissent le resultat precedent. Nous developpons dans la suite plusieurs applications de ce resultat. Nous montrons d'abord, comment construire des mesures de gibbs sur l'espace quotient / et sur un domaine fondamental de la relation d'equivalence , a partir de mesures de gibbs sur. Nous en deduisons une construction de mesures de gibbs sur l'espace des orbites et sur un domaine fondamental d'un groupe. Nous nous interessons ensuite a une loi gibbsienne p d'une collection infinie de processus de markov (x i) i zd en interaction, et nous montrons que sa mesure image par une transformation de dynkin est la loi gibbsienne de la collection infinie de processus de markov y = (y i) i zd = ((x i) i zd) en interaction. Enfin, par transformation de systemes gradients stochastiques, nous obtenons des resultats sur l'existence des solutions, leur unicite et des informations concernant les lois initiales reversibles, pour les dynamiques images, comme l'equation interface de ginzburg-landau.
19
Mang, Scott. "Changepoint analysis using Gibbs sampling." Thesis, Mang, Scott (2003) Changepoint analysis using Gibbs sampling. Honours thesis, Murdoch University, 2003. https://researchrepository.murdoch.edu.au/id/eprint/41249/.
Full text
20
Osterud, Erin Lee. "Gibbon classification : the issue of species and subspecies." PDXScholar, 1988. https://pdxscholar.library.pdx.edu/open_access_etds/3925.
Full textAbstract:
Gibbon classification at the species and subspecies levels has been hotly debated for the last 200 years. This thesis explores the reasons for this debate. Authorities agree that siamang, concolor, kloss and hoolock are species, while there is complete lack of agreement on lar, agile, moloch, Mueller's and pileated. The disagreement results from the use and emphasis of different character traits, and from debate on the occurrence and importance of gene flow.
21
Oddy, Carolyn. "The Gibbs Phenomenon and its Resolution." Thesis, Linnéuniversitetet, Institutionen för matematik (MA), 2015. http://urn.kb.se/resolve?urn=urn:nbn:se:lnu:diva-45850.
Full textAbstract:
It is well known that given an arbitrary continuous and periodic function f(x), it is possible to represent it as a Fourier series. However, attempting to approximate a discontinuous or non periodic function, using a Fourier series, yields very poor results. Large oscillations and overshoots appear around the points of discontinuity. Regardless of the number of terms that are included in the series, these overshoots do not disappear, they simply move closer to the point of discontinuity. This is known as the Gibbs phenomenon. In the 1990's David Gottlieb and Chi-Wang Shu introduced a new method, entitled the Gegenbauer procedure, which completely removes the Gibbs phenomenon. We will review their method, as well as present a number of examples to illustrate its effect. Going one step further we will discover that not all orthogonal polynomials may be treated equal in terms of this Gegenbauer procedure. When replacing Gegenbauer polynomials with Chebyshev or Legendre polynomials, it appears as though the inability to vary ultimately makes them ineffective.
22
Fay, TH, and PH Kloppers. "The Gibbs’ phenomenon for Fourier–Bessel series." International Journal of Mathematical Education in Science and Technology, 2003. http://encore.tut.ac.za/iii/cpro/DigitalItemViewPage.external?sp=1001984.
Full textAbstract:
Summary
The paper investigates the Gibbs’ phenomenon at a jump discontinuity for
Fourier–Bessel series expansions. The unexpected thing is that the Gibbs’
constant for Fourier–Bessel series appears to be the same as that for Fourier
series expansions. In order to compute the coefficients for Fourier–Bessel
functionsefficiently, several integral formulasare derived and the Struve
functions and their asymptotic expansions discussed, all of which significantly
ease the computations. Three numerical examples are investigated. Findings
suggest further investigations suitable for undergraduate research projects or
small student group investigations.
23
Krause, Andreas Eckhard. "Computerintensive statistische Methoden : Gibbs Sampling in Regressionsmodellen /." Stuttgart ; Jena : G. Fischer, 1994. http://bvbr.bib-bvb.de:8991/F?func=service&doc_library=BVB01&doc_number=006405906&line_number=0001&func_code=DB_RECORDS&service_type=MEDIA.
Full text
24
Stoehr, Julien. "Méthodes d'inférence statistique pour champs de Gibbs." Thesis, Montpellier, 2015. http://www.theses.fr/2015MONTS132/document.
Full textAbstract:
La constante de normalisation des champs de Markov se présente sous la forme d'une intégrale hautement multidimensionnelle et ne peut être calculée par des méthodes analytiques ou numériques standard. Cela constitue une difficulté majeure pour l'estimation des paramètres ou la sélection de modèle. Pour approcher la loi a posteriori des paramètres lorsque le champ de Markov est observé, nous remplaçons la vraisemblance par une vraisemblance composite, c'est à dire un produit de lois marginales ou conditionnelles du modèle, peu coûteuses à calculer. Nous proposons une correction de la vraisemblance composite basée sur une modification de la courbure au maximum afin de ne pas sous-estimer la variance de la loi a posteriori. Ensuite, nous proposons de choisir entre différents modèles de champs de Markov cachés avec des méthodes bayésiennes approchées (ABC, Approximate Bayesian Computation), qui comparent les données observées à de nombreuses simulations de Monte-Carlo au travers de statistiques résumées. Afin de pallier l'absence de statistiques exhaustives pour ce choix de modèle, des statistiques résumées basées sur les composantes connexes des graphes de dépendance des modèles en compétition sont introduites. Leur efficacité est étudiée à l'aide d'un taux d'erreur conditionnel original mesurant la puissance locale de ces statistiques à discriminer les modèles. Nous montrons alors que nous pouvons diminuer sensiblement le nombre de simulations requises tout en améliorant la qualité de décision, et utilisons cette erreur locale pour construire une procédure ABC qui adapte le vecteur de statistiques résumés aux données observées. Enfin, pour contourner le calcul impossible de la vraisemblance dans le critère BIC (Bayesian Information Criterion) de choix de modèle, nous étendons les approches champs moyens en substituant la vraisemblance par des produits de distributions de vecteurs aléatoires, à savoir des blocs du champ. Le critère BLIC (Block Likelihood Information Criterion), que nous en déduisons, permet de répondre à des questions de choix de modèle plus large que les méthodes ABC, en particulier le choix conjoint de la structure de dépendance et du nombre d'états latents. Nous étudions donc les performances de BLIC dans une optique de segmentation d'imagesDue to the Markovian dependence structure, the normalizing constant of Markov random fields cannot be computed with standard analytical or numerical methods. This forms a central issue in terms of parameter inference or model selection as the computation of the likelihood is an integral part of the procedure. When the Markov random field is directly observed, we propose to estimate the posterior distribution of model parameters by replacing the likelihood with a composite likelihood, that is a product of marginal or conditional distributions of the model easy to compute. Our first contribution is to correct the posterior distribution resulting from using a misspecified likelihood function by modifying the curvature at the mode in order to avoid overly precise posterior parameters.In a second part we suggest to perform model selection between hidden Markov random fields with approximate Bayesian computation (ABC) algorithms that compare the observed data and many Monte-Carlo simulations through summary statistics. To make up for the absence of sufficient statistics with regard to this model choice, we introduce summary statistics based on the connected components of the dependency graph of each model in competition. We assess their efficiency using a novel conditional misclassification rate that evaluates their local power to discriminate between models. We set up an efficient procedure that reduces the computational cost while improving the quality of decision and using this local error rate we build up an ABC procedure that adapts the summary statistics to the observed data.In a last part, in order to circumvent the computation of the intractable likelihood in the Bayesian Information Criterion (BIC), we extend the mean field approaches by replacing the likelihood with a product of distributions of random vectors, namely blocks of the lattice. On that basis, we derive BLIC (Block Likelihood Information Criterion) that answers model choice questions of a wider scope than ABC, such as the joint selection of the dependency structure and the number of latent states. We study the performances of BLIC in terms of image segmentation
25
Sake, Lekhya Sai. "ESTIMATION ON GIBBS ENTROPY FOR AN ENSEMBLE." CSUSB ScholarWorks, 2015. https://scholarworks.lib.csusb.edu/etd/264.
Full textAbstract:
In this world of growing technology, any small improvement in the present scenario would create a revolution. One of the popular revolutions in the computer science field is parallel computing. A single parallel execution is not sufficient to see its non-deterministic features, as same execution with the same data at different time would end up with a different path. In order to see how non deterministic a parallel execution can extend up to, creates the need of the ensemble of executions. This project implements a program to estimate the Gibbs Entropy for an ensemble of parallel executions. The goal is to develop tools for studying the non-deterministic feature of parallel code based on execution entropy and use these developed tools for current and future research.
26
Kerr, Christine. "Lewis Crassic Gibbon/James Leslie Mitchell : gender, sex and sexualities." Thesis, University of Sussex, 2002. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.249103.
Full textAbstract:
This thesis examines Lewis Grassic Gibbon/James Leslie Mitchell's promotion of the transformative power of the feminine and its offshoot of freer sexuality as a basis for new relationships among individuals, their societies and the world. As a result of Gibbon's revaluation of gender values, the feminine becomes identified as innate human "good," largely subsumed under, and set in opposition to, the evil perpetrated by the masculine historical process. Distinct from actual women, the feminine can be reclaimed by men too, affecting Gibbon's representations of both sexes. The feminine emerges as revolutionary - and not a conservative form of symbolism limiting women's subjectivity - in that it prepares the ground for a return to society and fuels both men and women with the power to challenge society's (masculine) values and institutions. A world-view structured around a gender dichotomy is nothing new. An overview of Gibbon's literary contemporaries, however, reveals that his prioritising of gender and sexual issues is unusual for a Scots male writer of the 1920sl1930s, although it does align him with female, feminist writers of his period (ch. I). Gibbon's early writing reconsiders stereotypes and archetypes of women/femininity but does not advance a practical programme for change (eh. 2). The influence ofDiffusionism, pronounced after 1930, manifests itself by portrayals of the male's re-connection to his "pre-civilisation" self through the feminine, allowing men and women together to renounce evils such as religion and masculine versions of history (eh. 3). Chapter 4 analyses various models that interact with factors such as race and sexual orientation to transcend gender disunity, although Gibbon's vision is occasionally marred by scepticism and blind-spots. His later work reveals a developing conviction that the individual- male or female - may have to lead the battle against evil, aiding transmission of the idea of "good." This, however, may lead to an overwriting of essential feminine values as is seen by the ending of A Scots Quair (chs. 5 & 6). In an analysis giving equal weight to most of his fiction, the thesis concludes that Gibbon's first step to solving civilization's malaise is a movement beyond polarities that make genders and sexes antagonistic, a ''third way," creating a rebirth of the individual and society when people re-awaken to the divinity in self and other and reconnect to the feminine. This movement, however, runs the risk of staying conjectural since actual measures for social change prove harder for Gibbon to delineate.
27
Ahsan, Md Farid. "Behavioural ecology of the hoolock gibbon (Hylobates hoolock) in Bangladesh." Thesis, University of Cambridge, 1994. https://www.repository.cam.ac.uk/handle/1810/283691.
Full text
28
Ebert, Rebecca. "Finding Laura Gibb /." Title page only, 2005. http://web4.library.adelaide.edu.au/theses/09AR/09are165.pdf.
Full text
29
Kakati, Kashmira. "Impact of forest fragmentation on the hoolock gibbon in Assam, India." Thesis, University of Cambridge, 2005. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.615270.
Full text
30
Schwabe, Jeremia [Verfasser], and Rolf [Akademischer Betreuer] Mülhaupt. "Gibbsit- und Aluminiumoxid-Einkristallnanoplättchen für PE-Nanokomposite, Mehrzentrenträgerkatalysatoren und selbstverstärkendes Polyethylen." Freiburg : Universität, 2015. http://d-nb.info/1119805759/34.
Full text
31
Yocom, Anna M. "Physical and Social Cognition in the White-handed Gibbon (Hylobates lar)." The Ohio State University, 2010. http://rave.ohiolink.edu/etdc/view?acc_num=osu1285083190.
Full text
32
Bagci, Gokhan B. "The Nonextensive Generalisation Of Boltzmann-gibbs Its Applications." Phd thesis, METU, 2005. http://etd.lib.metu.edu.tr/upload/3/12605845/index.pdf.
Full textAbstract:
This thesis analyzes the nonextensive generalization of Boltzmann-Gibbs statistics and study its applications to some physical models such as isotropic rigid and non-rigid rotators. The thesis will also try to show what kind of internal energy constraint must be chosen for entropy optimization in a mathematically consistent manner.
33
Chandgotia, Nishant. "Markov random fields, Gibbs states and entropy minimality." Thesis, University of British Columbia, 2015. http://hdl.handle.net/2429/52913.
Full textAbstract:
The well-known Hammersley-Clifford Theorem states (under certain conditions) that any Markov random field is a Gibbs state for a nearest neighbour interaction. Following Petersen and Schmidt we utilise the formalism of cocycles for the homoclinic relation and introduce "Markov cocycles", reparametrisations of Markov specifications. We exploit this formalism to deduce the conclusion of the Hammersley-Clifford Theorem for a family of Markov random fields which are outside the theorem's purview (including Markov random fields whose support is the d-dimensional "3-colored chessboard"). On the other extreme, we construct a family of shift-invariant Markov random fields which are not given by any finite range shift-invariant interaction.
The techniques that we use for this problem are further expanded upon to obtain the following results: Given a "four-cycle free" finite undirected graph H without self-loops, consider the corresponding 'vertex' shift, H ơm(Zd,H) denoted by X(H). We prove that X(H) has the pivot property, meaning that for all distinct configurations x,y ∈ X(H) which differ only at finitely many sites there is a sequence of configurations (x=x¹),x²,...,(xn =y) ∈ X(H) for which the successive configurations (xi,xi+1) differ exactly at a single site. Further if H is connected we prove that X(H) is entropy minimal, meaning that every shift space strictly contained in X(H) has strictly smaller entropy.
The proofs of these seemingly disparate statements are related by the use of the 'lifts' of the configurations in X(H) to their universal cover and the introduction of 'height functions' in this context.
Further we generalise the Hammersley-Clifford theorem with an added condition that the underlying graph is bipartite. Taking inspiration from Brightwell and Winkler we introduce a notion of folding for configuration spaces called strong config-folding to prove that if all Markov random fields supported on X are Gibbs with some nearest neighbour interaction so are Markov random fields supported on the "strong config-folds" and "strong config-unfolds" of X.Science, Faculty of
Mathematics, Department of
Graduate
34
Woodward, Christina Anna. "Philip Gibbs: war correspondent of a new dispensation." Thesis, Rhodes University, 1985. http://hdl.handle.net/10962/d1003126.
Full textAbstract:
The process of democratization which appeared in the nineteenth century was partly responsible for the emergence of a mass readership. It consisted of the new urban population which had its own tastes and interests, intellectual capacity and purchasing power. The popular press was firmly established by 1900 and it radically altered the scope and style of daily journalism in its attempt to speak in the language of the majority. Philip Gibbs was one of the prominent journalists between 1900 and 1914. His aspiration to become a war correspondent stemmed from the image of the war correspondent as a figure of romance and adventure, the consequence of the militarist spirit of the age and the licence which granted him freedom of movement. Inevitably, the war correspondent carne in conflict with the military which had not kept pace with democratization and sensed a challenge to itself and to national security. Censorship and restrictions on the war correspondent tightened, until major army reforms between 1901 and 1912 brought more cordial relations between the press and the military. When the Great War broke out in 1914 the co-operative atmosphere broke down as censorship was reinstated, more severely than before. It challenged the freedom of the press and the right of the people to know. Gibbs was determined that the people should have access to news from the front. He fought hard for that objective and was instrumental in the compromise reached between the military and the press when an officially recognized system was devised for press representation on the Western Front. The wisdom of such a move was shown by the success of Philip Gibbs' war correspondence, which had appeal to a mass readership in its own language and with subjects of interest to it.
35
Piao, Dongzhen. "Speeding Up Gibbs Sampling in Probabilistic Optical Flow." Research Showcase @ CMU, 2014. http://repository.cmu.edu/dissertations/481.
Full textAbstract:
In today’s machine learning research, probabilistic graphical models are used extensively to model complicated systems with uncertainty, to help understanding of the problems, and to help inference and predict unknown events. For inference tasks, exact inference methods such as junction tree algorithms exist, but they suffer from exponential growth of cluster size and thus is not able to handle large and highly connected graphs. Approximate inference methods do not try to find exact probabilities, but rather give results that improve as algorithm runs. Gibbs sampling, as one of the approximate inference methods, has gained lots of traction and is used extensively in inference tasks, due to its ease of understanding and implementation. However, as problem size grows, even the faster algorithm needs a speed boost to meet application requirement. The number of variables in an application graphical model can range from tens of thousands to billions, depending on problem domain. The original sequential Gibbs sampling may not return satisfactory result in limited time. Thus, in this thesis, we investigate in ways to speed up Gibbs sampling. We will study ways to do better initialization, blocking variables to be sampled together, as well as using simulated annealing. These are the methods that modifies the algorithm itself. We will also investigate in ways to parallelize the algorithm. An algorithm is parallelizable if some steps do not depend on other steps, and we will find out such dependency in Gibbs sampling. We will discuss how the choice of different hardware and software architecture will affect the parallelization result. We will use optical flow problem as an example to demonstrate the various speed up methods we investigated. An optical flow method tries to find out the movements of small image patches between two images in a temporal sequence. We demonstrate how we can model it using probabilistic graphical model, and solve it using Gibbs sampling. The result of using sequential Gibbs sampling is demonstrated, with comparisons from using various speed up methods and other optical flow methods.
36
Sidell, Ruth. "Regulation of messenger RNA stability in Lemna gibba." Thesis, University of Sheffield, 1997. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.267069.
Full text
37
Pang, Wan-Kai. "Modelling ordinal categorical data : a Gibbs sampler approach." Thesis, University of Southampton, 2000. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.323876.
Full text
38
Fair, Shannon Marie. "A Bayesian Meta-Analysis Using the Gibbs Sampler." UNF Digital Commons, 1998. http://digitalcommons.unf.edu/etd/87.
Full textAbstract:
A meta-analysis is the combination of results from several similar studies, conducted by different scientists, in order to arrive at a single, overall conclusion. Unlike common experimental procedures, the data used in a meta-analysis happen to be the descriptive statistics from the distinct individual studies.
In this thesis, we will consider two regression studies performed by two scientists. These studies have one common dependent variable, Y, and one or more independent common variables, X. A regression of Y on X with other independent variables is carried out on both studies. We will estimate the regression coefficients of X meta-analytically. After combining the two studies, we will derive a single regression model. There will be observations that one scientist witnesses and the other does not. The missing observations are considered parameters and are estimated using a method called Gibbs sampling.
39
Clarke, Esther A. E. "The vocalisations and anti-predatory behaviour of wild white-handed gibbons (Hylobates lar) in Khao Yai National Park, Thailand." Thesis, University of St Andrews, 2010. http://hdl.handle.net/10023/1688.
Full textAbstract:
The loud songs of gibbons (Hylobatidae) usually consist of a duet by the mated pair delivered each morning. These songs can transmit over a kilometre through dense forest habitat and therefore presumably play a role in long-distance communication. There is some evidence to suggest that gibbons use song in contexts other than their daily duets, such as predation, but these songs have not been well studied. Close- range communication is also relevant for gibbons, but these quieter calls have completely escaped any detailed observation. The responses of wild white-handed gibbons (Hylobates lar) to simulated visual and acoustic predators (tiger, clouded leopard, reticulated python and crested serpent eagle) were studied in Khao Yai National Park, Thailand to address the lack of empirical data about these important events. Little is known about gibbons’ anti- predatory behaviour in general, and simulated predator encounters provided an opportunity to investigate these responses as well. Results showed that gibbons used song as part of their anti-predator strategy and that subtle combinatorial changes were meaningful to conspecifics. They also showed marked behavioural changes in the short-term, and some evidence of longer-term changes as well. Quiet calls were also part of the gibbons’ response repertoire with the hoo call being particularly relevant. Hoos were used as a prelude to singing both normal duets and predator songs, but there were consistent differences between each context. Hoos were also delivered independently in a number of other contexts outside predation. When analysed, these hoos showed consistent contextual differences in a number of spectral parameters. Within the duet context, important contextual subtleties were evident also revealing a remarkable vocal plasticity. In addition, gibbons voluntarily attended to specific vocal elements of other gibbon duets, indicating that certain sequences are more pertinent than others. Results suggest both gibbon song and gibbon hoos are powerful communication tools that reliably reference external objects and events; this ability is also a critical feature of human language.
40
Becheikh, Samia. "Recherche sur l'écologie et le développement de "Pachypygus gibber" (Thorell, 1859), copépode Notodelphyidae associé à l'ascidie "Ciona intestinalis" (L. )." Montpellier 2, 1996. http://www.theses.fr/1996MON20240.
Full textAbstract:
Pachypygus gibber (thorell, 1859) est un copepode notodelphyiadae associe en mediterranee a l'ascidie ciona intestinalis (l. ). Cette espece est remarquable par l'existence de deux formes sexuelles males, qualifiees l'une de typique, l'autre d'atypique. L'apparition de l'une ou de l'autre des deux formes depend de facteurs environnementaux. L'analyse de quelques parametres ecologiques de cette association dans l'etang de thau, a montre que cet organisme presente une distribution agregee au sein de la population hote et au sein des individus hotes. Le sex ratio est biaise en faveur des femelles. L'accumulation de ce parasite n'affecte pas de facon significative ni la survie ni la croissance de la cione. Le cout de cette association est assez restreint. Une etude experimentale a permis de determiner la nature des facteurs epigenetiques influencant le dimorphisme male et la differenciation sexuelle chez ce copepode. Il s'agit de la ressource et du conjoint. En effet, le type de developpement de la larve infestante, elevee in vitro, depend de la quantite de nourriture presente dans le milieu. La determination du sexe depend egalement de la ressource mais surtout de la presence dans le milieu d'elevage d'un conjoint adulte ; la larve se developpe alors selon le sexe oppose. L'action du conjoint est limitee a une periode bien determinee du developpement de la larve. Une seconde etude experimentale a permis de mettre en evidence des phenomenes attractifs chez le copepodite 2 et le male atypique de p. Gibber. Si la larve infestante de ce parasite est attiree par la cione, le male atypique est en revanche attire par une femelle hebergee par une grande cione parasitee
41
Henning, Florian Benedikt [Verfasser], Christof [Gutachter] Külske, Peter [Gutachter] Eichelsbacher, and Sabine [Gutachter] Jansen. "Gibbs measures and gradient Gibbs measures on regular trees / Florian Benedikt Henning ; Gutachter: Christof Külske, Peter Eichelsbacher, Sabine Jansen ; Fakultät für Mathematik." Bochum : Ruhr-Universität Bochum, 2021. http://nbn-resolving.de/urn:nbn:de:hbz:294-83354.
Full text
42
Norman, Brian. "The influence of Switzerland on the life and writings of Edward Gibbon." Thesis, Open University, 1999. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.268257.
Full text
43
Betz, Volker. "Gibbs measures relative to Brownian motion and Nelson's model." [S.l. : s.n.], 2002. http://deposit.ddb.de/cgi-bin/dokserv?idn=964465647.
Full text
44
MAILLARD, GREGORY. "Chaines a liaisons completes et mesures de Gibbs unidimensionnelles." Phd thesis, Université de Rouen, 2003. http://tel.archives-ouvertes.fr/tel-00005285.
Full textAbstract:
On introduit un formalisme de mecanique statistique pour l'etude des processus stochastiques discrets (chaines) pour lesquels on prouve : (i) des proprietes generales de chaines extremales, incluant la trivialite de la tribu queue, les correlations a courtes portees, la realisation via des limites a volumes infinis et l'ergodicite, (ii) deux nouvelles conditions pour l'unicite de la chaine coherante, (iii) des resultats de perte de memoire et des proprietes de melange pour des chaines sous le regime de Dobrushin. On considere des systemes a alphabet fini, pouvant avoir une grammaire. On etablit des conditions pour qu'une chaine definisse une mesure de Gibbs et vice-versa. On discute de l'equivalence des criteres d'unicite pour les chaines et les champs et on etablit des bornes pour les taux de continuite des systemes respectifs de probabilites conditionnelles. On prouve un theoreme de (re)construction pour les specifications en partant de conditionnement sur un site.
45
Hörmann, Wolfgang, and Josef Leydold. "Monte Carlo Integration Using Importance Sampling and Gibbs Sampling." Department of Statistics and Mathematics, Abt. f. Angewandte Statistik u. Datenverarbeitung, WU Vienna University of Economics and Business, 2005. http://epub.wu.ac.at/1642/1/document.pdf.
Full textAbstract:
To evaluate the expectation of a simple function with respect to a complicated multivariate density Monte Carlo integration has become the main technique. Gibbs sampling and importance sampling are the most popular methods for this task. In this contribution we propose a new simple general purpose importance sampling procedure. In a simulation study we compare the performance of this method with the performance of Gibbs sampling and of importance sampling using a vector of independent variates. It turns out that the new procedure is much better than independent importance sampling; up to dimension five it is also better than Gibbs sampling. The simulation results indicate that for higher dimensions Gibbs sampling is superior. (author's abstract)Series: Preprint Series / Department of Applied Statistics and Data Processing
46
McFarlane, Ian. "An AIDS model of food demand using Gibbs sampling." Thesis, University of Reading, 2004. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.414580.
Full text
47
Kohatsu, Higa Arturo. "Estados de Gibbs para un sistema de partículas interactuantes." Pontificia Universidad Católica del Perú, 2013. http://repositorio.pucp.edu.pe/index/handle/123456789/97392.
Full text
48
Costa, Elias da. "Medidas de Gibbs e o Teorema de Aizenman-Higuchi." reponame:Repositório Institucional da UnB, 2013. http://repositorio.unb.br/handle/10482/14441.
Full textAbstract:
Dissertação (mestrado)—Universidade de Brasília, Instituto de Ciências Exatas, Departamento de Matemática, 2013.Submitted by Jaqueline Ferreira de Souza (jaquefs.braz@gmail.com) on 2013-10-30T09:26:40Z No. of bitstreams: 1 2013_EliasdaCosta_Parcial.pdf: 94443 bytes, checksum: 4d38e1280c4bf24fd8c8215d3de01b0d (MD5)
Approved for entry into archive by Guimaraes Jacqueline(jacqueline.guimaraes@bce.unb.br) on 2013-10-30T10:21:14Z (GMT) No. of bitstreams: 1 2013_EliasdaCosta_Parcial.pdf: 94443 bytes, checksum: 4d38e1280c4bf24fd8c8215d3de01b0d (MD5)
Made available in DSpace on 2013-10-30T10:21:14Z (GMT). No. of bitstreams: 1 2013_EliasdaCosta_Parcial.pdf: 94443 bytes, checksum: 4d38e1280c4bf24fd8c8215d3de01b0d (MD5)
Na primeira parte deste trabalho, apresentamos a teoria geral das medidas de Gibbs. A abordagem é baseada nas equações DLR e no formalismo termodinâmico. Em seguida, estudamos o modelo de Ising ferromagnético bidimensional. Mostramos que este modelo possui a propriedade forte de Markov e também algumas desigualdades de correlação, por exemplo a desigualdade de FKG. Por último provamos o Teorema de Aizenman-Higuchi o principal resultado desta dissertação. Este teorema sobre decomposição extremal foi provado independentemente, no inícios dos anos oitenta, por Michael Aizenman e Atsushi Higuchi, ambos baseados nos trabalhos de Lucio Russo. A prova dada aqui, devido a Aizenman, se baseia na investigação das simetrias dos espaços de configurações duplas e na aplicação sistemática da desigualdade de FKG e das equações DLR. _______________________________________________________________________________________ ABSTRACT
In the first part of this work, we present the general Gibbs measure theory. The approach is based on the DLR equations and the Thermodynamical Formalism. Next we study the ferromagnetic Ising model on the square lattice. We prove that this model satisfy the strong Markov property and also prove some correlation inequalities, as for example FKG. In the end we prove the Aizenman-Higuchi's theorem which is the main result of this master thesis. This theorem is about extremal decomposition and it was proved independently by Michael Aizenman and Atsushi Higuchi, both based on the work of Lucio Russo. The proof given here is due to Aizenman and is made by the investigation of the double configuration space symetries and systematic application of the FKG inequality and the DLR equations.
49
Maillard, Grégory. "Chaînes à liaisons complètes et mesures de Gibbs unidimensionnelles." Rouen, 2003. http://www.theses.fr/2003ROUES015.
Full textAbstract:
On introduit un formalisme de mécanique statistique pour l'étude des processus stochastiques discrets(chaînes) pour lesquels on prouve : (i) des propriétés générales de chaînes extrémales, incluant la trivialité de la tribu queue, les corrélations à courtes portées, la réalisation via des limites à volumes infinis et l'ergodicité, (ii) deux nouvelles conditions pour l'unicité de la chaîne cohérente, (iii) des résultats de perte de mémoire et des propriétés de mélange pour des chaînes sous le régime de Dobrushin. On considère des systèmes à alphabet fini, pouvant avoir une grammaire. On établit des conditions pour qu'une chaîne définisse une mesure de Gibbs et vice-versa. On discute de l'équivalence des critères d'unicité pour les chaînes et les champs et on établit des bornes pour les taux de continuité des systèmes respectifs de probabilités conditionnelles. On prouve un théorème de (re)construction pour les spécifications en partant de conditionnement sur un siteWe introduce an statistical mechanical formalism for the study of discrete-time stochastic processes (chains) with which we prove: (i) General properties of extremal chains, including triviality on the tail sigma-algebra, short-range correlations, realization via infinite-volume limits and ergodicity. (ii) Two new sufficient conditions for the uniqueness of the consistent chain. (iii) Results on loss of memory and mixing properties for chains in the Dobrushin regime. We discuss the relationship between chains and one-dimensional Gibbs measures. We consider finite-alphabet systems, possibly with a grammar. We establish conditions for a chain to define a Gibbs measure and vice versa. We discuss the equivalence of uniqueness criteria for chains and fields and we establish bounds for the continuity rates of the respective systems of finite-volume conditional probabilities. We prove a (re)construction theorem for specifications starting from single-site conditioning
50
Bryant, J. V. "Developing a conservation evidence-base for the Critically Endangered Hainan gibbon (Nomascus hainanus)." Thesis, University College London (University of London), 2014. http://discovery.ucl.ac.uk/1434514/.
Full textAbstract:
The Critically Endangered Hainan gibbon (Nomascus hainanus) is the world’s rarest ape, with a single population of an estimated 25 individuals surviving in approximately 15 km² of suboptimal forest within Bawangling National Nature Reserve, Hainan, China. The existing biological evidence-base for the species is inadequate for conservation planning, precluding evaluation of appropriate recovery actions. I derived comprehensive new baseline data on Hainan gibbon ecology, behaviour and genetics to clarify the species’ biology and population status, and inform urgently required conservation management for the species. Rigorous re-evaluation of Hainan gibbon spatial requirements indicated the species’ home range is much smaller than previously estimated (c. 1.5 km²) and in line with closely related Nomascus species in similar ecological conditions. Molecular assessment of the genetic status of the surviving population within the context of the species’ historical genetic diversity revealed that the Hainan gibbon has suffered a significant decline in genetic diversity following its past population bottleneck. The current population also shows a high level of relatedness and male-biased offspring sex ratio. Predictive models examining reportedly ‘anomalous’ Hainan gibbon ecological and behavioural traits within a phylogenetic framework indicated that large, polygynous groups may be evolutionarily characteristic for the species, but home range requirements are influenced by both intrinsic factors and current extrinsic conditions. Finally, Population Viability Analysis demonstrated that the species is highly likely to become extinct in the near future without active management, and that multiple actions mitigating extrinsic threats, enhancing habitat carrying capacity and improving survivorship will be required to reduce extinction risk. Together, these findings suggest that landscape-level management actions and intensive manipulation of the population may be necessary to safeguard the future of the Hainan gibbon. This research also has wider implications for improved understanding of gibbon ecology and conservation of species of extreme rarity.