1,721,059 research outputs found
The limits of Nečiporuk’s method and the power of programs over monoids taken from small varieties of finite monoids
Get PDFCotutelle avec l'École Normale Supérieure de Cachan, Université Paris-Saclay.Cette thèse porte sur des minorants pour des mesures de complexité liées à des sous-classes de la classe P de langages pouvant être décidés en temps polynomial par des machines de Turing. Nous considérons des modèles de calcul non uniformes tels que les programmes sur monoïdes et les programmes de branchement.
Notre première contribution est un traitement abstrait de la méthode de Nečiporuk pour prouver des minorants, indépendamment de toute mesure de complexité spécifique. Cette méthode donne toujours les meilleurs minorants connus pour des mesures telles que la taille des programmes de branchements déterministes et non déterministes ou des formules avec des opérateurs booléens binaires arbitraires ; nous donnons une formulation abstraite de la méthode et utilisons ce cadre pour démontrer des limites au meilleur minorant obtenable en utilisant cette méthode pour plusieurs mesures de complexité. Par là, nous confirmons, dans ce cadre légèrement plus général, des résultats de limitation précédemment connus et exhibons de nouveaux résultats de limitation pour des mesures de complexité auxquelles la méthode de Nečiporuk n’avait jamais été appliquée.
Notre seconde contribution est une meilleure compréhension de la puissance calculatoire des programmes sur monoïdes issus de petites variétés de monoïdes finis. Les programmes sur monoïdes furent introduits à la fin des années 1980 par Barrington et Thérien pour généraliser la reconnaissance par morphismes et ainsi obtenir une caractérisation en termes de semi-groupes finis de NC^1 et de ses sous-classes. Étant donné une variété V de monoïdes finis, on considère la classe P(V) de langages reconnus par une suite de programmes de longueur polynomiale sur un monoïde de V : lorsque l’on fait varier V parmi toutes les variétés de monoïdes finis, on obtient différentes sous-classes de NC^1, par exemple AC^0, ACC^0 et NC^1 quand V est respectivement la variété de tous les monoïdes apériodiques finis, résolubles finis et finis. Nous introduisons une nouvelle notion de docilité pour les variétés de monoïdes finis, renforçant une notion de Péladeau. L’intérêt principal de cette notion est que quand une variété V de monoïdes finis est docile, nous avons que P(V) contient seulement des langages réguliers qui sont quasi reconnus par morphisme par des monoïdes de V. De nombreuses questions ouvertes à propos de la structure interne de NC^1 seraient réglées en montrant qu’une variété de monoïdes finis appropriée est docile, et, dans cette thèse, nous débutons modestement une étude exhaustive de quelles variétés de monoïdes finis sont dociles. Plus précisément, nous portons notre attention sur deux petites variétés de monoïdes apériodiques finis bien connues : DA et J. D’une part, nous montrons que DA est docile en utilisant des arguments de théorie des semi-groupes finis. Cela nous permet de dériver une caractérisation algébrique exacte de la classe des langages réguliers dans P(DA). D’autre part, nous montrons que J n’est pas docile. Pour faire cela, nous présentons une astuce par laquelle des programmes sur monoïdes de J peuvent reconnaître beaucoup plus de langages réguliers que seulement ceux qui sont quasi reconnus par morphisme par des monoïdes de J. Cela nous amène à conjecturer une caractérisation algébrique exacte de la classe de langages réguliers dans P(J), et nous exposons quelques résultats partiels appuyant cette conjecture. Pour chacune des variétés DA et J, nous exhibons également une hiérarchie basée sur la longueur des programmes à l’intérieur de la classe des langages reconnus par programmes sur monoïdes de la variété, améliorant par là les résultats de Tesson et Thérien sur la propriété de longueur polynomiale pour les monoïdes de ces variétés.This thesis deals with lower bounds for complexity measures related to subclasses of the class P of languages that can be decided by Turing machines in polynomial time. We consider non-uniform computational models like programs over monoids and branching programs.
Our first contribution is an abstract, measure-independent treatment of Nečiporuk’s method for proving lower bounds. This method still gives the best lower bounds known on measures such as the size of deterministic and non-deterministic branching programs or formulæ with arbitrary binary Boolean operators; we give an abstract formulation of the method and use this framework to prove limits on the best lower bounds obtainable using this method for several complexity measures. We thereby confirm previously known limitation results in this slightly more general framework and showcase new limitation results for complexity measures to which Nečiporuk’s method had never been applied.
Our second contribution is a better understanding of the computational power of programs over monoids taken from small varieties of finite monoids. Programs over monoids were introduced in the late 1980s by Barrington and Thérien as a way to generalise recognition by morphisms so as to obtain a finite-semigroup-theoretic characterisation of NC^1 and its subclasses. Given a variety V of finite monoids, one considers the class P(V) of languages recognised by a sequence of polynomial-length programs over a monoid from V: as V ranges over all varieties of finite monoids, one obtains different subclasses of NC^1, for instance AC^0, ACC^0 and NC^1 when V respectively is the variety of all finite aperiodic, finite solvable and finite monoids. We introduce a new notion of tameness for varieties of finite monoids, strengthening a notion of Péladeau. The main interest of this notion is that when a variety V of finite monoids is tame, we have that P(V) does
only contain regular languages that are quasi morphism-recognised by monoids from V. Many open questions about the internal structure of NC^1 would be settled by showing that some appropriate variety of finite monoids is tame, and, in this thesis, we modestly start an exhaustive study of which varieties of finite monoids are tame. More precisely, we focus on two well-known small varieties of finite aperiodic monoids: DA and J. On the one hand, we show that DA is tame using finite-semigroup-
theoretic arguments. This allows us to derive an exact algebraic characterisation of the class of regular languages in P(DA). On the other hand, we show that J is not tame. To do this, we present a trick by which programs over monoids from J can recognise much more regular languages than only those that are quasi morphism-recognised by monoids from J. This brings us to conjecture an exact algebraic characterisation of the class of regular languages in P(J), and we lay out some partial results that support this conjecture. For each of the varieties DA and J, we also exhibit a program-length-based hierarchy within the class of languages recognised by programs over monoids from the variety, refining Tesson and Thérien’s results on the polynomial-length property for monoids from those varieties
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Reasoning about distributed relational data and query evaluation
Get PDFLarge data sets are often stored distributedly to increase the reliability of systems and the efficiency of query evaluation in them. While some query operators -- like selections and projections -- are intrinsically conform with parallel evaluation, others -- like joins -- demand specific distribution patterns.
For relational databases, a common approach to evaluate queries in parallel relies on the use of rather simple distribution patterns for binary joins and the computation of the query result according to some query plan, operator by operator. Often, this requires the redistribution of large intermediate results (possibly larger than the input and/or output) and thus may lead to unnecessary long processing times. Thus, especially in the last decade, more elaborate distribution patterns that depend on the whole query have been studied and shown to allow more efficient query evaluation in several cases by reducing the amount of communication between servers.
Ameloot et al. have described a setting where query evaluation is studied for a broad range of distribution patterns. Their work focuses on problems to decide whether a query can be evaluated correctly under a given distribution pattern. More particularly, they have considered two problems: "parallel correctness", where the pattern is specified explicitly, and "parallel-correctness transfer", where the pattern is known to be appropriate for another query.
This thesis comprises the author's contributions to the complexity-theoretical investigation of these problems for conjunctive queries (and extensions thereof). These contributions complement the main characterisations and some additional complexity results by Ameloot et al. Furthermore, this thesis contains some new characterisations for "polarised" queries.
Via the characterisations, parallel correctness and parallel-correctness transfer can be translated into questions on the co-occurrences of certain facts, induced by the query, on some server. Such questions and others can be modelled by "distribution dependencies", a variant of the well-known tuple- and equality-generating dependencies.
Modelling via these constraints allows a more general description of distribution patterns in distributed relational data.
The third contribution of this thesis is the study of the implication problem for distribution dependencies, providing lower and upper bounds for some fragments
Variations on the Author
Get PDF“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Get PDFWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Énumération des requêtes et graphes nulle-part dense
No full textLes travaux présentés dans ma thèse se situent à l’interface entre complexité, algorithmique et logique. Plus particulièrement, on s’intéresse à la complexité d'évaluation de requêtes.Plus précisément, étant donné G un graphe fini. Une requête q définit un sous ensemble de k-uplets de sommets de G que l'on note q(G). On appelle k l'arité de q et on se essaye alors d'effectuer efficacement les taches suivantes :1) décider si l'ensemble q(G) est vide ou non.2) décider si un k-uplet donné appartient à l'ensemble des solutions q(G).3) calculer le nombre de solutions.4) énumérer les éléments de q(G).En ce qui concerne la 4ème tache, un algorithme qui va énumérer les solutions sera décomposé en deux parties. La première est appelé le pré-calcul et sert à préparer l’énumération. Idéalement cette étape de requière qu’un temps linéaire en la taille du graphe. La deuxième étape est ensuite l’énumération des solutions. Le temps nécessaire pour obtenir une nouvelle solution est appelé le délai. Idéalement on souhaite que le délai de dépende pas de la taille du graphes mais uniquement de la taille de la requête. On parle alors d’énumération à délai constant après pré-calcul linéaire.Au début de cette thèse, une grand part des interrogations au sujet des classes de graphes pour lesquelles une énumération à délai constant serait possible semblait se trouver au niveau des classes de graphes nulle-part dense. Le résultat principal de cette thèse est de montrer qu’il est possible d’énumérer les solutions des requêtes du premier ordre sur les graphes nulle-part dense avec un délai constant après un pré-calcul pseudo linéaire.The topic of my thesis lies between complexity, algorithmic and logic. In particular, we are interested in the complexity of evaluating query.More precisely, given G a finite graph. A query q defines a subset of k-tuples of vertices of G that we note q(G). We call k the arity of q and we then try to efficiently perform the following tasks:1) decide whether the set q G) is empty.2) decide whether a given k-tuplet belongs to the set of solutions q(G).3) calculate the number of solutions.4) enumerate the elements of q(G).Regarding the 4th task, an algorithm that will enumerate the solutions can be decomposed into two steps. The first is called preprocessing and is used to prepare the enumeration. Ideally this step only requires a time linear in the size of the graph. The second step is the enumeration properly speaking. The time needed to get a new solution is called the delay. Ideally we want the delay to not depend on the size of the graph but only on the size of the query. We then talk about constant delay enumeration after linear preprocessing.At the beginning of this thesis, a large part of the interrogations about classes of graphs for which a constant delay enumeration is possible seemed to be located around the classes of nowhere dense graph
On Invariant Formulae of First-Order Logic with Numerical Predicates
Get PDFDiese Arbeit untersucht ordnungsinvariante Formeln der Logik erster Stufe
(FO) und einiger ihrer Erweiterungen, sowie andere eng verwandte Konzepte der endlichen Modelltheorie. Viele Resultate der endlichen Modelltheorie nehmen an, dass Strukturen mit einer Einbettung ihres Universums in ein Anfangsstück der natürlichen Zahlen ausgestattet sind. Dies erlaubt es, beliebige Relationen (z.B. die lineare Ordnung) und Operationen (z.B. Addition, Multiplikation) von den natürlichen Zahlen auf solche Strukturen zu übertragen.
Die resultierenden Relationen auf den endlichen Strukturen werden als numerische Prädikate bezeichnet. Werden numerische Prädikate in Formeln verwendet, beschränkt man sich dabei häufig auf solche Formeln, deren Wahrheitswert auf endlichen Strukturen invariant unter Änderungen der Einbettung der Strukturen ist. Wenn das einzige verwendete numerische Prädikat eine lineare Ordnung ist, spricht man beispielsweise von ordnungsinvarianten Formeln. Die Resultate dieser Arbeit können in drei Teile unterteilt werden.
Der erste Teil betrachtet die Lokalitätseigenschaften von FO-Formeln mit Modulo-Zählquantoren, die beliebige numerische Prädikate invariant nutzen.
Der zweite Teil betrachtet FO-Sätze, die eine lineare Ordnung samt der zugehörigen Addition auf invariante Weise nutzen, auf endlichen Bäumen. Es wird gezeigt, dass diese dieselben regulären Baumsprachen definieren, wie FO-Sätze ohne numerische Prädikate mit bestimmten Kardinalitätsprädikaten. Für den Beweis wird eine algebraische Charakterisierung der in dieser Logik definierbaren Baumsprachen durch Operationen auf Bäumen entwickelt.
Der dritte Teil der Arbeit beschäftigt sich mit der Ausdrucksstärke und der Prägnanz
von FO und Erweiterungen von FO auf Klassen von Strukturen beschränkter Baumtiefe.This thesis studies the concept of order-invariance of formulae of first-order logic (FO)
and some of its extensions as well as other closely related concepts from finite model theory.
Many results in finite model theory assume that structures are equipped with an
embedding of their universe into an initial segment of the natural numbers. This allows
to transfer arbitrary relations (e.g. linear order) and operations (e.g. addition, multiplication)
on the natural numbers to structures. The arising relations on the structures are
called numerical predicates. If formulae use these numerical predicates, it is often desirable
to consider only such formulae whose truth value in finite structures is invariant under changes to the embeddings of the structures. If the numerical predicates include only a linear order, such formulae are called order-invariant. We study the effect of the invariant use of different kinds of numerical predicates on the expressive power of FO and extensions thereof. The results of this thesis can be divided into three parts.
The first part considers the locality and non-locality properties of formulae of FO with
modulo-counting quantifiers which may use arbitrary numerical predicates in an invariant way. The second part considers sentences of FO which may use a linear
order and the corresponding addition in an invariant way and obtains a characterisation of the regular finite tree languages which can be defined by such sentences: these are the same tree languages which are definable by FO-sentences without numerical predicates with certain cardinality predicates. For the proof, we obtain a characterisation of the tree languages definable in this logic in terms of algebraic operations on trees.
The third part compares the expressive power and the succinctness of different ex-
tensions of FO on structures of bounded tree-depth
Dispelling the Myths Behind First-author Citation Counts
Get PDFWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
- …
