Dada uma palavra w e um grupo G, suponha que o conjunto Gw pode ser coberto por finitos subgrupos cÃclicos. à verdade que w(G) tambÃm pode ser coberto por finitos subgrupos cÃclicos? Nesta dissertaÃÃo mostraremos que a resposta à positiva para a palavra comutador. / Given a word w and a group G, suppose that the set can be Gw covered by finite cyclic subgroups. It is true that w(G) can also be covered by finite cyclic subgroups? This dissertation will show that the answer is positive for the word switch.

Les variétés de Deligne-Lusztig associées à un élément de Coxeter, dites variétés de Coxeter et notées $YY(dot{c})$, sont des variétés candidates à réaliser l'équivalence dérivée demandée dans la conjecture de Broué. Cette conjecture implique qu'une telle variété doit avoir une cohomologie disjointe et donne également la description de l'algèbre d'endomorphismes associée. Dans le cas des groupes linéaires, nous décrivons la cohomologie des variétés de Coxeter et en déduisons que celles-ci vérifient bien les propriétés impliquées par la conjecture de Broué. Pour ce faire, nous montrons qu'il est possible d'appliquer un résultat de og transitivitéfg permettant de se ramener à des variétés de Coxeter og plus petitesfg et nous utilisons ensuite un résultat établi par Lusztig sur des variétés notées $XX(c)$, obtenues comme des quotients des variétés $YY(dot{c})$ par des groupes finis. Enfin, dans une dernière partie, la description de la cohomologie des variétés de Coxeter nous permet d'obtenir un lien entre la cohomologie de la compactification $overline{YY}(dot{c})$ et celle de la compactification $overline{XX}(c)$. / Deligne-Lusztig varieties associated to Coxeter elements, or more simply Coxeter Varieties denoted by $YY(dot{c})$, are good candidates to realize the derived equivalence needed for the Broué's conjecture. The conjecture implies that the varieties should have disjoint cohomology as well as gives a description of the endomorphisms algebra.For linear groups, we describe the cohomology of the Coxeter varieties and hence show that it agrees with the conditions implied by Broué's conjecture. To do so, we prove it is possible to apply a og transitivityfg result allowing us to restrict to og smallerfg Coxeter varieties. Then, we apply a result obtained by Lusztig on varieties $XX(c)$, which are quotient varieties of $YY(dot{c})$ by some finite groups.In the last part of the thesis, we use the description of the cohomology of Coxeter varieties to connect the cohomology of the compactification $overline{YY}(dot{c})$ and the cohomology of the compactification $overline{XX}(c)$.

This thesis demonstrates an alternative, concise but informative, method for representing group elements, which will prove particularly useful for the sporadic groups. It explains the theory behind symmetric presentations, and describes the algorithm for working with elements represented in this manner.

In the Introduction of the Local Structure Theorem for finite groups with a large p-subgroup [MSS16] Meierfrankenfeld, Stellmacher and Stroth write: “It is obvious that one cannot get any information about M and its action on YM without discussing in one way or another the embedding of M into G. But a priori, it is not clear at all what type of embedding properties one should study and how they would help to get this information”. Here G is a finite group and M a cleverly chosen subgroup, with YM being a canonically determined elementary abelian, normal p-subgroup of M. On the one side, an answer to the question contained in the above citation is given by the result proven by Meierfrankenfeld, Stellmacher and Stroth; on the other side, one may ask how much of the group G is necessary for such result and how much is lost or changed by retaining only the plocal information of G, i.e. the information carried by the normalizers of non-trivial p-subgroups. In some way, which will hopefully become more clear and specific by the end of this Introduction, this dissertation can be seen as attempt to answer such question: we replace the group G with localities, as defined in 2.2.7, our choice of structures encoding precisely the p-local information of G, and look for an analogous of the Local Structure Theorem. In order to provide enough context to understand the statement and reach of [MSS16] as well as reasons and ideas behind the development of localities, we need a long historical digression and a detailed enough description of the strategy that was successful in proving the Local Structure Theorem. For the readers’ convenience we accordingly split our Introduction.

In this thesis we study gradings on blocks of group algebras. The motivation to study gradings on blocks of group algebras and their transfer via derived and stable equivalences originates from some of the most important open conjectures in representation theory, such as Broue’s abelian defect group conjecture. This conjecture predicts the existence of derived equivalences between categories of modules. Some attempts to prove Broue’s conjecture by lifting stable equivalences to derived equivalences highlight the importance of understanding the connection between transferring gradings via stable equivalences and transferring gradings via derived equivalences. The main idea that we use is the following. We start with an algebra which can be easily graded, and transfer this grading via derived or stable equivalence to another algebra which is not easily graded. We investigate the properties of the resulting grading. In the first chapter we list the background results that will be used in this thesis. In the second chapter we study gradings on Brauer tree algebras, a class of algebras that contains blocks of group algebras with cyclic defect groups. We show that there is a unique grading up to graded Morita equivalence and rescaling on an arbitrary basic Brauer tree algebra. The third chapter is devoted to the study of gradings on tame blocks of group algebras. We study extensively the class of blocks with dihedral defect groups. We investigate the existence, positivity and tightness of gradings, and we classify all gradings on these blocks up to graded Morita equivalence. The last chapter deals with the problem of transferring gradings via stable equivalences between blocks of group algebras. We demonstrate on three examples how such a transfer via stable equivalences is achieved between Brauer correspondents, where the group in question is a TI group.

Dans cette thèse, on s'intéresse au groupe de Brauer non ramifié des espaces homogènes à stabilisateur non connexe et à ses applications arithmétiques. On développe notamment différentes formules de nature algébrique et/ou arithmétique permettant de calculer explicitement, tant sur un corps fini que sur un corps de caractéristique 0, la partie algébrique du groupe de Brauer non ramifié d'un espace homogène G\G' sous un groupe linéaire G' semi-simple simplement connexe à stabilisateur fini G, le tout en donnant des exemples de calculs que l'on peut faire avec ces formules. Pour ce faire, on démontre au préalable (à l'aide d'un théorème de Gabber sur les altérations) un résultat décrivant la partie de torsion première à p du groupe de Brauer non ramifié d'une variété V lisse et géométriquement intègre sur un corps fini ou sur un corps global de caractéristique p au moyen de l'évaluation des éléments de Br(V) sur ses points locaux. Les formules pour un stabilisateur fini sont ensuite généralisées au cas d'un stabilisateur G quelconque via une réduction de la cohomologie galoisienne du groupe G à celle d'un certain sous-quotient fini. Enfin, pour K un corps global et G un K-groupe fini résoluble, on démontre sous certaines hypothèses sur une extension déployant G que l'espace homogène V:=G\G' avec G' un K-groupe semi-simple simplement connexe vérifie l'approximation faible (ces hypothèses assurant la nullité du groupe de Brauer non ramifié algébrique). On utilise une version plus précise de ce résultat pour démontrer ensuite le principe de Hasse pour des espaces homogènes X sous un K-groupe G' semi-simple simplement connexe à stabilisateur géométrique fini et résoluble, sous certaines hypothèses sur le K-lien défini par X. / This thesis studies the unramified Brauer group of homogeneous spaces with non connected stabilizer and its arithmetic applcations. In particular, we develop different formulas of algebraic and/or arithmetic nature allowing an explicit calculation, both over a finite field and over a field of characteristic 0, of the algebraic part of the unramified Brauer group of a homogeneous space G\G' under a semisimple simply connected linear group G' with finite stabilizer G. We also give examples of the calculations that can be done with these formulas. For achieving this goal, we prove beforehand (using a theorem of Gabber on alterations) a result describing the prime-to-p torsion part of the unramified Brauer group of a smooth and geometrically integral variety V over a global field of characteristic p or over a finite field by evaluating the elements of Br(V) at its local points. The formulas for finite stabilizers are later generalised to the case where the stabilizer G is any linear algebraic group using a reduction of the Galois cohomology of the group G to that of a certain finite subquotient.Finally, for a global field K and a finite solvable K-group G, we show under certain hypotheses concerning the extension splitting G that the homogeneous space V:=G\G' with G' a semi-simple simply connected K-group has the weak approximation property (the hypotheses ensuring the triviality of the unramified algebraic Brauer group). We use then a more precise version of this result to prove the Hasse principle forhomogeneous spaces X under a semi-simple simply connected K-group G' with finite solvable geometric stabilizer, under certain hypotheses concerning the K-kernel (or K-lien) defined by X.

Submitted by Bruno Caldeira Carlotti de Souza null (brunoccarlotti@gmail.com) on 2017-03-02T01:45:21Z No. of bitstreams: 1 Dissertação - Bruno C. C. de Souza.pdf: 1030573 bytes, checksum: e3dd1e43953565236359b6d10831067c (MD5) / Approved for entry into archive by LUIZA DE MENEZES ROMANETTO (luizamenezes@reitoria.unesp.br) on 2017-03-07T18:32:31Z (GMT) No. of bitstreams: 1 souza_bcc_me_sjrp.pdf: 1030573 bytes, checksum: e3dd1e43953565236359b6d10831067c (MD5) / Made available in DSpace on 2017-03-07T18:32:31Z (GMT). No. of bitstreams: 1 souza_bcc_me_sjrp.pdf: 1030573 bytes, checksum: e3dd1e43953565236359b6d10831067c (MD5) Previous issue date: 2017-02-23 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / O objetivo principal deste trabalho é apresentar detalhadamente um estudo sobre um critério, que aparece na referência Coincidence for maps of spaces with finite group action de D. L. Gonçalves, J. Jaworowski, P. L. Q. Pergher e A. Volovikov, para a existência de (H,G)-coincidências de aplicações cujo contradomínio é um CW-complexo finito Y de dimensão k e cujo domínio é um espaço X paracompacto, Hausdorff, conexo e localmente conexo por caminhos e munido de uma G-ação livre, de modo que exista um inteiro m tal que os grupos i-dimensionais de homologia de X sejam triviais nas dimensões 0<i<m e a cohomologia (m+1)-dimensional de G não seja trivial. Para a realização deste estudo foram necessários alguns resultados da teoria de cohomologia de grupos finitos, com ênfase em grupos de cohomologia periódica segundo a teoria de cohomologia de Tate, alguns resultados da teoria de fibrados e algumas noções da teoria de sequências espectrais cohomológicas. / The mais objective of this work is to present in detail a study about a criterion, which appears in the reference Coincidence for maps of spaces with finite group actions by D. L. Gonçalves, J. Jaworowski, P. L. Q. Pergher and A. Volovikov, for existence of (H,G)-coincidences of maps into a finite CW-complex Y with dimension k and whose domain is a paracompact, Hausdorff, connected and locally pathconnected space X with a free action of G, in a way that there exists an integer m such that the ith-homology group of X is trivial for 0<i<m and the (m+1)th-cohomology group of G is nontrivial. For the study of this criterion were needed some results of the theory of cohomology of finite groups, with emphasis on groups with periodic cohomology according to Tate cohomology theory, some results of the theory of fiber bundles and some notions of the theory of cohomological spectral sequences.

Dans la premier partie de cette thèse, nous étudions la structure des sous-groupes approximatifs dans les groupes metabéliens (groupes résolubles de classe de résolubilité 2) et montrons que si A est un tel sous-groupe K approximatif, il est K^⁰(r) contrôlée (au sens du Tao) par un groupe nilpotent où $ r désigne le rang de $ G=Fit (G) et Fit (G) $ est le sous-groupe de fitting de G. La deuxième partie est consacrée à l'étude de la croissance des ensembles dans GLn(Fq) où Fq est un corps fini. Nous montrons une borne sur le diamètre (par rapport à n'importe quel système des générateurs) pour tous sous-groupes simples finis de ce groupe. Si G est un groupe fini simple de type Lie de rang n, et son corps de base est de taille borné, le diamètre du graphe du Cayley Gamma (G;S) serait borné par exp (O (n (log n) ^ 3)) . Si la taille du corps fini Fq n'est pas borné, notre méthode donne une borne de q ^ {O (n ( log nq) ^ 3) pour le diamètre.Dans la troisième partie nous nous sommes intéressés à la croissance des ensembles dans les boucles de Moufang commutatifs. Ceux-ci sont les boucles commutatifs respectant les identités de Moufang mais sans être (nécessairement) associatifs. Nous montrons que, si les tailles des ensembles des associateurs sont bornées alors la croissance des sous-structures approximatifs dans ces boucles est similaire à celle des groupes ordinaires. De cette façon dans le cadre des boucles de moufang commutatifs finiment engendré on a un théorème de structure pour ses sous-boucles approximatifs.Mots-clefs -sous-groupes approximatifs, groupes résolubles, diamètres des groupes, boucles de moufang commutatifs. / In the first part of this thesis, we study the structure of approximate subgroups inside metabelian groups (solvable groups of derived length 2) and show that if A is such a K-approximate subgroup, then it is K^(O(r)) controlled (in the sense of Tao) by a nilpotent group where r denotes the rank of G=Fit(G) and Fit(G) is the fitting subgroup of G.The second part is devoted to the study of growth of sets inside GLn(Fq) , where we show a bound on the diameter (with respect to any set of generators) for all finite simple subgroups of this group. What we have is - if G is a finite simple group of Lie type with rank n, and its base field has bounded size, then the diameter of the Cayley graph C(G; S) would be bounded by exp(O(n(logn)^3)). If the size of the base field Fq is not bounded then our method gives a bound of q^(O(n(log nq)3)) for the diameter.In the third part we are interested in the growth of sets inside commutative Moufang loops which are commutative loops respecting the moufang identities but without (necessarily)being associative. For them we show that if the sizes of the associator sets are bounded then the growth of approximate substructures inside these loops is similar to those in ordinary groups. In this way for the subclass of finitely generated commutative moufang loops we have a classification theorem of its approximate subloops.

Ma thèse se situe à l'intersection de \textit {la théorie des groupes géométrique} et \textit{la théorie des groupes mesurée}. Une question majeure dans la théorie des groupes géométrique est d'étudier la classe de quasi-isométrie (QI) et la classe d'équivalence mesurée (ME) d'un groupe, respectivement. $L^p$-équivalence mesurée est une relation d'équivalence qui est définie en ajoutant des contraintes géométriques avec d'équivalence mesurée. En plus, QI est une condition géométrique. Il est une question naturelle, si deux groupes sont QI et ME, si elles sont $L^p$-ME pour certains $p>0$. Dans mon premier article, en collaboration avec R. Tessera, nous répondons négativement à cette question pour $p\geq 1$, montrant que l'extension centrale canonique d'un groupe surface de genre plus élevé ne sont pas $L^1$-ME pour le produit direct de ce groupe de surface avec $\mathbb{Z}$ (alors qu'ils sont à la fois quasi-isométrique et équivalente mesurée).Dans mon deuxième papier, j'ai observé un lien général entre la géométrie des expandeurs, defini comme une séquence des quotients finis ( l'espace de boîte) d'un groupe finiment engendré, et les propriétés mesurée theorique du groupe. Plus précisément, je l'ai prouvé que si deux <<espaces de boîte>> sont quasi-isométrique, les groupes correspondants doivent être <<mesurée équivalente uniformément >>, une notion qui combine à la fois QI et ME. Je prouve aussi une version de ce résultat pour le plongement grossière, ce qui permet de distinguer plusieurs classe des expandeurs. Par exemple, je montre que les expandeurs associé à $SL(m, \mathbb{Z})$ ne grossièrement plongent à les expandeurs associés à $SL_n(\mathbb{Z})$ si $m>n$. / My thesis lies at the intersection of \textit{geometric group theory} and \textit{measured group theory}. A major question in geometric group theory is to study the quasi-isometry (QI) class and the measure equivalence (ME) class of a group, respectively. $L^p$-measure equivalence is an equivalence relation which is defined by adding some geometric constraints with measure equivalence. Besides, quasi-isometry is a geometric condition. It is a natural question if two groups are QI and ME, whether they are $L^p$-ME for some $p>0$. In my first paper, together with R. Tessera, we answer this question negatively for $p\geq 1$, showing that the canonical central extension of a surface group of higher genus is not $L^1$-ME to the direct product of this surface group with $\mathbb{Z}$ (while they are both quasi-isometric and measure equivalent). In my second paper, I observed a general link between the geometry of expanders arising as a sequence of finite quotients (box space) of a finitely generated group, and the measured theoretic properties of the group. More precisely, I proved that if two box spaces' are quasi-isometric, then the corresponding groups must be `uniformly measure equivalent', a notion that combines both quasi-isometry and measure equivalence. I also prove a version of this result for coarse embedding, allowing to distinguish many classes of expanders. For instance, I show that the expanders associated to $SL(m,\mathbb{Z})$ do not coarsely embed inside the expanders associated to $SL_n(\mathbb{Z}$ if $m>n$.

Made available in DSpace on 2014-07-29T16:02:19Z (GMT). No. of bitstreams: 1 Dissertacao Marcio Lima.pdf: 1529346 bytes, checksum: c6a80a13d55b40203c44877c4cdeb1f4 (MD5) Previous issue date: 2011-06-27 / A be an elementary abelian group of order q2, where q a prime number. In this paper we will study the influence of centering on the structure of automorphism groups profinitos in this sense if A acting as a coprime group of automorphisms on a group profinito G and CG(a) is periodic for each a 2 A#, then we will show that G is locally finite. It will be demonstrated also the case where A acts as a group of automorphisms of a group pro-p of G / Sejam A um grupo abeliano elementar de ordem q2, onde q um número primo. Neste trabalho estudamos a influência dos centralizadores de automorfismos na estrutura dos grupos profinitos, neste sentido se A age como um grupo de automorfismos coprimos sobre um grupo profinito G e que CG(a) é periódico para cada a 2 A#, então mostraremos que G é localmente finito. Será demonstrado também o caso onde A age como um grupo de automorfismos sobre um grupo pro-p de G.