Papers

This page hopefully stays being under construction. You can read some questions here.


On conjugate chains of subgroups in residually finite groups

 

We show that conjugate descending chains of finite index subgroups can have different intersections in a residually finite discrete group. In fact, we can obtain an arbitrary countable set of residually finite groups as the intersection set of a conjugacy class of such chains.

 


with L. Babai, Finite groups of  uniform logarithmic diameter, to appear in Israel J. of Math. PDF

 

We provide an example for a sequence of finite groups $G_n$ such that all $G_n$ can be generated by 2 elements and the diameter of any Cayley graph of $G_n$ is $Clog(G_n)$. This answers a question of Lubotzky.

 


On the probability of satisfying a word in a group, to appear in J. of Group Theory PDF

 

We show that for any finite group $G$ and for any $d$ there exists a word $w\in F_{d}$ such that a $d$-tuple in $G$ satisfies $w$ if and only if it generates a solvable subgroup. In particular, if $G$ itself is not solvable, then it cannot be obtained as a quotient of the one relator group $F_{d}/<w>$.
As a corollary, the probability that a word is satisfied in a fixed non-solvable group can be made arbitrarily small, answering a question of Alon Amit.

 


with Y. Glasner, Generic groups acting on trees

 

Let A be the automorphism group of a regular tree and let G be the subgroup generated by k random elements in A. We show that almost surely G is free of rank k and every nonidentity element of G has 0 or 2 fixpoints on the boundary of the tree. Also, for k > 1 almost surely the closure of G is discrete, compact, equal to A or has index 2 in A. All four possibilities have infinite measure in $A^k$.

 


Group laws and free subgroups in topological groups, Bull. London Math. Soc. 37 (2005), no. 4, 525-534 ArXiV

 

We prove that a permutation group in which different finite sets have different stabilizers cannot satisfy any group law. For locally compact topological groups with this property we show that almost all finite subsets of the group generate free subgroups.
We derive consequences of these theorems on Thompson's group F, weakly branch groups, automorphism groups of regular trees and profinite groups with alternating composition factors of unbounded degree.


with B. Virág, Dimension and randomness in groups acting on rooted trees, J. Amer. Math. Soc. 18 (2005), no. 1, 157-192 ArXiV

 

We explore the structure of the p-adic automorphism group Γ(p) of the infinite rooted regular tree. We determine the asymptotic order of a typical element, answering an old question of Turan.
We initiate the study of a general dimension theory of groups acting on rooted trees. We describe the relationship between dimension and other properties of groups such as solvability, existence of dense free subgroups and the normal subgroup structure. We show that subgroups of Γ(p) generated by three random elements are full-dimensional and that there exist finitely generated subgroups of arbitrary dimension. Specifically, our results solve an open problem of Shalev and answer a question of Sidki.


with N. Nikolov and B. Szegedy, Congruence subgroup growth of arithmetic groups in positive characteristic, Duke Math J. (2003), no. 2, 367—383 PDF

 

We prove a new uniform bound for subgroup growth of a Chevalley group G over the local ring F[[t]] and also over local pro-p rings of higher Krull dimension. This is applied to the determination of congruence subgroup growth of arithmetic groups over global fields of positive characteristic. In particular we obtain that the subgroup growth of SL_n(F_p[t]) (n > 2) is of type $n^{\log{n}}$. This was one of the main problems left open by Alex Lubotzky in his

article.

 

The essential tool for proving the results is the use of graded Lie algebras. We sharpen Lubotzky`s bounds on subgroup growth via a result on subspaces of a Chevalley Lie algebra L over a finite field. The proof of this theorem is by algebraic geometry and can be modified to obtain a lower bound on the codimension of proper Lie subalgebras of L.


with T. Keleti, Shuffle the plane, Proc. Amer. Math. Soc. 130 (2002), no. 2, 549-553 PS

 

We prove that every permutation of the real plane can be obtained as the composition of a fixed number of simple transformations of the form (x,y) ® (x,y+f(x)) and (x,y) ® (x+g(y),y) where f and g are arbitrary real functions.

 


Symmetric groups as products of Abelian subgroups, Bull. London Math. Soc. 34 (2002), no. 4, 451-456 PS

 

We prove that the full symmetric group over any infinite set is the product of finitely many Abelian subgroups. In fact, 289 subgroups suffice.

We also obtain sharp bounds on the minimal number k such that the finite symmetric group S(n) is the product of k Abelian subgroups. Using this, we prove that S(n) is the product of $72n^{1/2}(\log n)^{3/2}$ cyclic subgroups.


with A. Lubotzky and L. Pyber, Bounded generation and linear groups, Int. J. Alg. Comp., 13 (2003), no. 4, 401-413 PS

 

A group is called boundedly generated (BG) if it is the set-theoretic product of finitely many cyclic subgroups. We show that a BG group has only abelian by finite images in positive characteristic representations.

We use this to reprove and generalise Rapinchuk's theorem by showing that a BG group with the FAb property has only finitely many irreducible representations in any given dimension over any field. We also give a structure theorem for the profinite completion of such a group.

On the other hand, we exhibit boundedly generated profinite FAb groups which do not satisfy this structure theorem.


with L. Pyber, Decomposing finite linear groups into the product of cyclic groups

 

Let K be a field. We show that if G is a finite completely reducible subgroup of GL(n,K) then G can be obtained as a product of Cn cyclic subgroups where C is an absolute constant.


Symmetric presentations of Abelian groups, Proc. Amer. Math. Soc. 131 (2003), no. 1, 17-20 PDF

 

We characterise the abelianisation of a group, that has a presentation for which the set of relations is invariant under the full symmetric group acting on the set of generators. This improves a result of Emerson.