Structure learning of context-specific graphical models

Abstract The ultimate problem considered in this thesis is modeling a high-dimensional joint distribution over a set of discrete variables. For this purpose, we consider classes of context-specific graphical models and the main emphasis is on learning the structure of such models from data. Traditional graphical models compactly represent a joint distribution through a factorization justi ed by statements of conditional independence which are encoded by a graph structure. Context-speci c independence is a natural generalization of conditional independence that only holds in a certain context, speci ed by the conditioning variables. We introduce context-speci c generalizations of both Bayesian networks and Markov networks by including statements of context-specific independence which can be encoded as a part of the model structures. For the purpose of learning context-speci c model structures from data, we derive score functions, based on results from Bayesian statistics, by which the plausibility of a structure is assessed. To identify high-scoring structures, we construct stochastic and deterministic search algorithms designed to exploit the structural decomposition of our score functions. Numerical experiments on synthetic and real-world data show that the increased exibility of context-specific structures can more accurately emulate the dependence structure among the variables and thereby improve the predictive accuracy of the models.

Proof theory for hybrid(ised) logics

Hybridisation is a systematic process along which the characteristic features of hybrid logic, both at the syntactic and the semantic levels, are developed on top of an arbitrary logic framed as an institution. In a series of papers this process has been detailed and taken as a basis for a speci cation methodology for recon gurable systems. The present paper extends this work by showing how a proof calculus (in both a Hilbert and a tableau based format) for the hybridised version of a logic can be systematically generated from a proof calculus for the latter. Such developments provide the basis for a complete proof theory for hybrid(ised) logics, and thus pave the way to the development of (dedicated) proof support.

Garbage collection optimization for non uniform memory access architectures

Cache-coherent non uniform memory access (ccNUMA) architecture is a standard design pattern for contemporary multicore processors, and future generations of architectures are likely to be NUMA. NUMA architectures create new challenges for managed runtime systems. Memory-intensive applications use the system’s distributed memory banks to allocate data, and the automatic memory manager collects garbage left in these memory banks. The garbage collector may need to access remote memory banks, which entails access latency overhead and potential bandwidth saturation for the interconnection between memory banks. This dissertation makes five significant contributions to garbage collection on NUMA systems, with a case study implementation using the Hotspot Java Virtual Machine. It empirically studies data locality for a Stop-The-World garbage collector when tracing connected objects in NUMA heaps. First, it identifies a locality richness which exists naturally in connected objects that contain a root object and its reachable set— ‘rooted sub-graphs’. Second, this dissertation leverages the locality characteristic of rooted sub-graphs to develop a new NUMA-aware garbage collection mechanism. A garbage collector thread processes a local root and its reachable set, which is likely to have a large number of objects in the same NUMA node. Third, a garbage collector thread steals references from sibling threads that run on the same NUMA node to improve data locality. This research evaluates the new NUMA-aware garbage collector using seven benchmarks of an established real-world DaCapo benchmark suite. In addition, evaluation involves a widely used SPECjbb benchmark and Neo4J graph database Java benchmark, as well as an artificial benchmark. The results of the NUMA-aware garbage collector on a multi-hop NUMA architecture show an average of 15% performance improvement. Furthermore, this performance gain is shown to be as a result of an improved NUMA memory access in a ccNUMA system. Fourth, the existing Hotspot JVM adaptive policy for configuring the number of garbage collection threads is shown to be suboptimal for current NUMA machines. The policy uses outdated assumptions and it generates a constant thread count. In fact, the Hotspot JVM still uses this policy in the production version. This research shows that the optimal number of garbage collection threads is application-specific and configuring the optimal number of garbage collection threads yields better collection throughput than the default policy. Fifth, this dissertation designs and implements a runtime technique, which involves heuristics from dynamic collection behavior to calculate an optimal number of garbage collector threads for each collection cycle. The results show an average of 21% improvements to the garbage collection performance for DaCapo benchmarks.

Problems in number theory and hyperbolic geometry

In the first part of this thesis we generalize a theorem of Kiming and Olsson concerning the existence of Ramanujan-type congruences for a class of eta quotients. Specifically, we consider a class of generating functions analogous to the generating function of the partition function and establish a bound on the primes ℓ for which their coefficients c(n) obey congruences of the form c(ℓn + a) ≡ 0 (mod ℓ). We use this last result to answer a question of H.C. Chan. In the second part of this thesis [S2] we explore a natural analog of D. Calegari’s result that there are no hyperbolic once-punctured torus bundles over S^1 with trace field having a real place. We prove a contrasting theorem showing the existence of several infinite families of pairs (−χ, p) such that there exist hyperbolic surface bundles over S^1 with trace field of having a real place and with fiber having p punctures and Euler characteristic χ. This supports our conjecture that with finitely many known exceptions there exist such examples for each pair ( −χ, p).

Increased hippocampal excitability and impaired spatial memory function in mice lacking VGLUT2 selectively in neurons defined by tyrosine hydroxylase promoter activity

Three populations of neurons expressing the vesicular glutamate transporter 2 (Vglut2) were recently described in the A10 area of the mouse midbrain, of which two populations were shown to express the gene encoding, the rate-limiting enzyme for catecholamine synthesis, tyrosine hydroxylase (TH).One of these populations (‘‘TH– Vglut2 Class1’’) also expressed the dopamine transporter (DAT) gene while one did not ("TH–Vglut2 Class2"), and the remaining population did not express TH at all ("TH-Vglut2-only"). TH is known to be expressed by a promoter which shows two phases of activation, a transient one early during embryonal development, and a later one which gives rise to stable endogenous expression of the TH gene. The transient phase is, however, not specific to catecholaminergic neurons, a feature taken to advantage here as it enabled Vglut2 gene targeting within all three A10 populations expressing this gene, thus creating a new conditional knockout. These knockout mice showed impairment in spatial memory function. Electrophysiological analyses revealed a profound alteration of oscillatory activity in the CA3 region of the hippocampus. In addition to identifying a novel role for Vglut2 in hippocampus function, this study points to the need for improved genetic tools for targeting of the diversity of subpopulations of the A10 area

Naps in school can enhance the duration of declarative memories learned by adolescents

Sleep helps the consolidation of declarative memories in the laboratory, but the pro-mnemonic effect of daytime naps in schools is yet to be fully characterized. While a few studies indicate that sleep can indeed benefit school learning, it remains unclear how best to use it. Here we set out to evaluate the influence of daytime naps on the duration of declarative memories learned in school by students of 10–15 years old. A total of 584 students from 6th grade were investigated. Students within a regular classroom were exposed to a 15-min lecture on new declarative contents, absent from the standard curriculum for this age group. The students were then randomly sorted into nap and non-nap groups. Students in the nap group were conducted to a quiet room with mats, received sleep masks and were invited to sleep. At the same time, students in the non-nap group attended regular school classes given by their usual teacher (Experiment I), or English classes given by another experimenter (Experiment II). These 2 versions of the study differed in a number of ways. In Experiment I (n = 371), students were pre-tested on lecture-related contents before the lecture, were invited to nap for up to 2 h, and after 1, 2, or 5 days received surprise tests with similar content but different wording and question order. In Experiment II (n = 213), students were invited to nap for up to 50 min (duration of a regular class); surprise tests were applied immediately after the lecture, and repeated after 5, 30, or 110 days. Experiment I showed a significant ∼10% gain in test scores for both nap and non-nap groups 1 day after learning, in comparison with pre-test scores. This gain was sustained in the nap group after 2 and 5 days, but in the non-nap group it decayed completely after 5 days. In Experiment II, the nap group showed significantly higher scores than the non-nap group at all times tested, thus precluding specific conclusions. The results suggest that sleep can be used to enhance the duration of memory contents learned in school.

Radiative Behaviour of Gd2O3:Er/Yb nanoparticles under near infrared illumination

Upconverting nanoparticles have attracted much attention in science recently, specifically in view of medical and biological applications such as live imaging of cell temperatures or cancer treatment. The previously studied system of gadolinium oxide nanorods co-doped with erbium and ytterbium and decorated with different number densities of gold nanoparticles has been studied. So far, these particles have been proven as efficient nanothermometers in a temperature range from 300 up to 2000 K. In this work, a more detailed study on the morphological and radiative behaviour of these particles has been conducted. It was found that the laser power threshold for the onset of the black body radiation decreases strongly with the increase in the gold concentration. The temperature of the onset itself seems to remain approximately constant. The heating efficiency was determined to increase significantly with the gold concentration. The morphological study revealed that the temperature at the black body radiation threshold was not enough to induce any significant transformation in neither the nanorods nor the gold nanoparticles, as was expected from comparison with literature. However, significant changes in radiative properties and the morphology were detected for powders that underwent strong laser heating until the emission of brightly visible black body radiation.

Idiosyncratic psychological aspects in entrepreneurship

Entrepreneurship is having the courage to transform an idea in reality and with it achieve personal, nancial and recognition satisfaction. The psychological ability to handle failure has proven essential in success. We analysed the importance of idiosyncratic psychological aspects in the success of entrepreneurs through an observational study accompanying 20 entrepreneurs from the idea presentation phase to company incorporation. During the observation period 4 distinct psychological phases of the entrepreneurs were observed, being it possible to describe them as follows: absorption of information and knowledge; application of the gathered knowledge to their speci c cases; frustration generated by criticism, namely from investors who didn’t recognise the value of their projects; realism and implementation of the project. Having passed more than 6 months after the analysis period, one can verify that the entrepreneurs who have travelled the 4 phases are today developing their projects being that the remaining ones are in a similar situation as at the end of the initial two months. Conclusion: The ability to cope with frustration and rejection is a determinant factor in the success of the entrepreneur. The ability to learn from rejection, more than resilience help the entrepreneur to proceed.

Contributions to gait recognition using multiple views

Esta tesis se centra en la identificación de personas a través de la forma de caminar. El problema del reconocimiento del paso ha sido tratado mediante diferentes enfoques, en los dominios 2D y 3D, y usando una o varias vistas. Sin embargo, la dependencia con respecto al punto de vista, y por tanto de la trayectoria del sujeto al caminar sigue siendo aún un problema abierto. Se propone hacer frente al problema de la dependencia con respecto a la trayectoria por medio de reconstrucciones 3D de sujetos caminando. El uso de reconstrucciones varias ventajas que cabe destacar. En primer lugar, permite explotar una mayor cantidad de información en contraste con los métodos que extraen los descriptores de la marcha a partir de imágenes, en el dominio 2D. En segundo lugar, las reconstrucciones 3D pueden ser alineadas a lo largo de la marcha como si el sujeto hubiera caminado en una cinta andadora, proporcionando así una forma de analizar el paso independientemente de la trayectoria seguida. Este trabajo propone tres enfoques para resolver el problema de la dependencia a la vista: 1. Mediante la utilización de reconstrucciones volumétricas alineadas. 2. Mediante el uso de reconstrucciones volumétricas no alineadas. 3. Sin usar reconstrucciones. Se proponen además tres tipos de descriptores. El primero se centra en describir el paso mediante análisis morfológico de los volúmenes 3D alineados. El segundo hace uso del concepto de entropa de la información para describir la dinámica del paso humano. El tercero persigue capturar la dinámica de una forma invariante a rotación, lo cual lo hace especialmente interesante para ser aplicado tanto en trayectorias curvas como rectas, incluyendo cambios de dirección. Estos enfoques han sido probados sobre dos bases de datos públicas. Ambas están especialmente diseñadas para tratar el problema de la dependencia con respecto al punto de vista, y por tanto de la dependencia con respecto a la trayectoria. Los resultados experimentales muestran que para el enfoque basado en reconstrucciones volumétricas alineadas, el descriptor del paso basado en entropa consigue los mejores resultados, en comparación con métodos estrechamente relacionados del Estado del Arte actual. No obstante, el descriptor invariante a rotación consigue una tasa de reconocimiento que supera a los métodos actuales sin requerir la etapa previa de alineamiento de las reconstrucciones 3D.

Thematic patterning in English and Spanish: contrastive annotation of a bilingual newspaper corpus for liguistic and computational applications

Thematization is recognized as a fundamental phenomenon in the construction of messages and texts by di erent linguistic schools. This location within a text privileges the elements that guide the reader in the orientation and interpretation of discourse at di erent levels. Thematizing a linguistic unit by locating it in the rst-initial position of a clause, paragraph, or text, confers upon it a special status: a signal of the organizational strategy which characterizes di erent text types playing a role as a variable in the distinction of registers, text types and genres. However, in spite of the importance of the study of thematization for message and textual structuring, to date there are no linguistic studies that have undertook the task of validating its aspects in a comparative manner, either for linguistic or computational purposes. This study, therefore, lls a research gap by implementing a methodology based on contrastive corpus annotation, which allows to empirically validate aspects of the phenomenon of Thematization in English and Spanish, it also seeks to develop a bilingual English-Spanish comparable corpus of newspaper texts automatically annotated with thematic features at clausal and discourse levels. The empirically validated categories (Thematic Field and its elements: Textual Theme, Interpersonal Theme, PreHead and Head) are used to annotate a larger corpus of three newspaper genres news reports, editorials and letters to the editor in terms of thematic choices. This characterization, reveals interesting results, such as the use of genre-speci c strategies in thematic position. In addition, the thesis investigates the possibility to automate the annotation of thematic features in the bilingual corpus through the development of a set of JAVA rules implemented in GATE. It also shows the e cacy of this method in comparison with the manual annotation results...

Evolution of developmental gene regulatory networks in echinoids

Developmental gene regulatory networks (dGRNs) are assemblages of regulatory genes that direct embryonic development of animal body plans and their morpho-logical structures. dGRNs exhibit recursively-wired circuitry that is encoded in the genome and executed during development. Alteration to the regulatory architecture of dGRNs causes variation in developmental programs both during the development of an individual organism and during the evolution of an individual lineage. The ex-planatory power of these networks is best exemplified by the global dGRN directing early development of the euechinoid sea urchin Strongylocentrotus purpuratus. This network consists of numerous regulatory genes engaging in hundreds of genomic regulatory transactions that collectively direct the delineation of early embryonic domains and the specification of cell lineages. Research on closely-related euechi-noid sea urchins, e.g. Lytechinus variegatus and Paracentrotus lividus, has revealed marked conservation of dGRN architecture in echinoid development, suggesting little appreciable alteration has occurred since their divergence in evolution at least 90 million years ago (mya).

We sought to test whether this observation extends to all sea urchins (echinoids) and undertook a systematic analysis of over 50 regulatory genes in the cidaroid sea urchin Eucidaris tribuloides, surveing their regulatory activity and function in a sea urchin that diverged from euechinoid sea urchins at least 268 mya. Our results revealed extensive alterations have occurred to all levels of echinoid dGRN archi-tecture since the cidaroid-euechinoid divergence. Alterations to mesodermal sub-circuits were particularly striking, including functional di˙erences in specification of non-skeletogenic mesenchyme (NSM), skeletogenic mesenchyme (SM), and en-domesodermal segregation. Specification of endomesodermal embryonic domains revealed that, while their underlying network circuitry had clearly diverged, regu-latory states established in pregastrular embryos of these two groups are strikingly similar. Analyses of E. tribuloides specification leading to the estab-lishment of dorsal-ventral (aboral-oral) larval polarity indicated that regulation of regulatory genes expressed in mesodermal embryonic domains had incurred significantly more alterations than those expressed in endodermal and ectodermal domains. Taken together, this study highlights the ability of dGRN architecture to buffer extensive alterations in the evolution and early development of echinoids and adds further support to the notion that alterations can occur at all levels of dGRN architecture and all stages of embryonic development.

The specific shapes of gender imbalance in scientific authorships : a network approach

Gender differences in collaborative research have received little at- tention when compared with the growing importance that women hold in academia and research. Unsurprisingly, most of bibliomet- ric databases have a strong lack of directly available information by gender. Although empirical-based network approaches are often used in the study of research collaboration, the studies about the influence of gender dissimilarities on the resulting topological outcomes are still scarce. Here, networks of scientific subjects are used to characterize patterns that might be associated to five categories of authorships which were built based on gender. We find enough evidence that gen- der imbalance in scientific authorships brings a peculiar trait to the networks induced from papers published in Web of Science (WoS) in- dexed journals of Economics over the period 2010-2015 and having at least one author affiliated to a Portuguese institution. Our re- sults show the emergence of a specific pattern when the network of co-occurring subjects is induced from a set of papers exclusively au- thored by men. Such a male-exclusive authorship condition is found to be the solely responsible for the emergence that particular shape in the network structure. This peculiar trait might facilitate future network analyses of research collaboration and interdisciplinarity.

Metals and linear alkylbenzene sulphonate as inhibitors of the algae Pseudokirchneriella subcapitata acid phosphatase activity.

Sewage sludge applied to soils as a fertilizer often contains metals and linear alkylbenzene sulphonate (LAS) as contaminants. These pollutants can be transported to the aquatic environment where they can alter the phosphatase activity in living organisms. The acid phosphatase of algae plays important roles in metabolism such as decomposing organic phosphate into free phosphate and autophagic digestive processes. The order of in vitro inhi- bition of Pseudokirchneriella subcapitata acid phosphatase at the highest concentration tested was LAS[Hg2? = Al 3?[Se4? = Pb2?[Cd2?. A non-competitive inhibi- tion mechanism was obtained for Hg2? (Ki = 0.040 mM) and a competitive inhibition for LAS (Ki = 0.007 mM). In vivo studies with treated algae cultures showed that the inhibition of specific activity was observed in algae exposed during 7 days, in contrast to short term (24 h) treatments with both these chemicals. Our results suggest that the inhibition parameters in vitro did not markedly differ between the two chemicals. On the other hand, in vivo evaluations showed strong differences between both pollu- tants regarding the concentration values and the degree of response.

The influence of organic amendments on soil aggregate stability from semiarid sites

Restoring the native vegetation is the most effective way to regenerate soil health. Under these conditions, vegetation cover in areas having degraded soils may be better sustained if the soil is amended with an external source of organic matter. The addition of organic materials to soils also increases infiltration rates and reduces erosion rates; these factors contribute to an available water increment and a successful and sustainable land management. The goal of this study was to analyze the effect of various organic amendments on the aggregate stability of soils in afforested plots. An experimental paired-plot layout was established in southern of Spain (homogeneous slope gradient: 7.5%; aspect: N170). Five amendments were applied in an experimental set of plots: straw mulching; mulch with chipped branches of Aleppo Pine (Pinus halepensis L.); TerraCotten hydroabsobent polymers; sewage sludge; sheep manure and control. Plots were afforested following the same spatial pattern, and amendments were mixed with the soil at the rate 10 Mg ha-1. The vegetation was planted in a grid pattern with 0.5 m between plants in each plot. During the afforestation process the soil was tilled to 25 cm depth from the surface. Soil from the afforested plots was sampled in: i) 6 months post-afforestation; ii) 12 months post-afforestation; iii) 18 months post-afforestation; and iv) 24 months post-afforestation. The sampling strategy for each plot involved collection of 4 disturbed soil samples taken from the surface (0–10 cm depth). The stability of aggregates was measured by wet-sieving. Regarding to soil aggregate stability, the percentage of stable aggregates has increased slightly in all the treatments in relation to control. Specifically, the differences were recorded in the fraction of macroaggregates (≥ 0.250 mm). The largest increases have been associated with straw mulch, pinus mulch and sludge. Similar results have been registered for the soil organic carbon content. Independent of the soil management, after six months, no significant differences in microaggregates were found regarding to the control plots. These results showed an increase in the stability of the macroaggregates when soil is amended with sludge, pinus mulch and straw much. This fact has been due to an increase in the number cementing agents due to: (i) the application of pinus, straw and sludge had resulted in the release of carbohydrates to the soil; and thus (ii) it has favored the development of a protective vegetation cover, which has increased the number of roots in the soil and the organic contribution to it.

Impacto de las características de la red eléctrica en canales MIMO PLC domésticos

The recent release of indoor Power Line Communications (PLC) specifications with Multiple-Input Multiple-Output (MIMO) capabilities has significantly increased the bit rates achieved in these channels. However, the performance reached by the use of these methods may differ from one location to another due to the heterogeneous nature of the domestic power grid. In this work, a closer look at the relation between channel performance and the power grid cabling is taken. To that end, some channel features like attenuation, spatial correlation and capacity are analyzed by means of a set of 50 simulated channel topologies in the frequency band from 1 to 80 MHz.