arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Author

Tim Austin

68 papers · 1957–2025

1966: 01967: 01968: 01969: 01970: 01971: 01972: 01973: 01974: 01975: 01976: 01977: 01978: 01979: 01980: 01981: 01982: 01983: 01984: 01985: 01986: 01987: 01988: 01989: 01990: 01991: 01992: 01993: 01994: 01995: 01996: 01997: 01998: 01999: 02000: 02001: 02002: 02003: 02004: 02005: 12006: 52007: 22008: 72009: 112010: 52011: 42012: 22013: 102014: 22015: 32016: 42017: 32018: 22019: 02020: 12021: 12022: 02023: 12024: 22025: 1
1966peak 11 in 20092025

Frequent co-authors

Papers

showing 1–25 of 68 RSS

On exchangeable random variables and the statistics of large graphs and hypergraphs

Tim Austin

math.PR · · cited 136× · arXiv ↗ · more like this ·

De Finetti's classical result of [18] identifying the law of an exchangeable family of random variables as a mixture of i.i.d. laws was extended to structure theorems for more complex notions of exchangeability by Aldous [1,2,3], Hoover [41,42], Kallenberg [44] and Kingman [47]. On the other hand, such exchangeable laws were first related to questions from combinatorics in an independent analysis by Fremlin and Talagrand [29], and again more recently in Tao [62], where they appear as a natural proxy for the `leading order statistics' of colourings of large graphs or hypergraphs. Moreover, this relation appears implicitly in the study of various more bespoke formalisms for handling `limit objects' of sequences of dense graphs or hypergraphs in a number of recent works, including Lovász and Szegedy [52], Borgs, Chayes, Lovász, Sós, Szegedy and Vesztergombi [17], Elek and Szegedy [24] and…

On the norm convergence of non-conventional ergodic averages

Tim Austin

Discrete Mathematics and Combinatorics · Ergodic Theory and Dynamical Systems · · cited 93× · DOI ↗ · more like this ·

Abstract We offer a proof of the following non-conventional ergodic theorem: If T i :ℤ r ↷( X ,Σ, μ ) for i =1,2,…, d are commuting probability-preserving ℤ r -actions, ( I N ) N ≥1 is a Følner sequence of subsets of ℤ r , ( a N ) N ≥1 is a base-point sequence in ℤ r and f 1 , f 2 ,…, f d ∈ L ∞ ( μ ) then the non-conventional ergodic averages converge to some limit in L 2 ( μ ) that does not depend on the choice of ( a N ) N ≥1 or ( I N ) N ≥1 . The leading case of this result, with r =1 and the standard sequence of averaging sets, was first proved by Tao, following earlier analyses of various more special cases and related results by Conze and Lesigne, Furstenberg and Weiss, Zhang, Host and Kra, Frantzikinakis and Kra and Ziegler. While Tao’s proof rests on a conversion to a finitary problem, we invoke only techniques from classical ergodic theory, so giving a new proof of his result.

On the norm convergence of nonconventional ergodic averages

Tim Austin

math.DS · · cited 92× · arXiv ↗ · more like this ·

We offer a generalization of the recent result of Tao (building on earlier results of Conze and Lesigne, Furstenberg and Weiss, Zhang, Host and Kra, Frantzikinakis and Kra and Ziegler) that the nonconventional ergodic averages associated to an arbitrary number of commuting probability-preserving transformations always converge to some limit in L^2. We prove the corresponding result for a collection of commuting actions of a larger discrete Abelian group, and gives convergence that is uniform in the start-point of the averages. While Tao's proof rests on a conversion to a finitary problem, we invoke only techniques from classical ergodic theory, so giving a new proof of his result.

Testability and repair of hereditary hypergraph properties

Tim Austin, Terence Tao

Computational Theory and Mathematics · Random Structures and Algorithms · · cited 47× · DOI ↗ · more like this ·

Abstract Recent works of Alon–Shapira (A characterization of the (natural) graph properties testable with one‐sided error. Available at: http://www.math.tau.ac.il/∼nogaa/PDFS/heredit2.pdf ) and Rödl–Schacht (Generalizations of the removal lemma, Available at: http://www.informatik.hu‐berlin.de/∼schacht/pub/preprints/gen_removal.pdf ) have demonstrated that every hereditary property of undirected graphs or hypergraphs is testable with one‐sided error; informally, this means that if a graph or hypergraph satisfies that property “locally” with sufficiently high probability, then it can be perturbed (or “repaired”) into a graph or hypergraph which satisfies that property “globally.” In this paper we make some refinements to these results, some of which may be surprising. In the positive direction, we strengthen the results to cover hereditary properties of multiple directed polychromatic…

On the testability and repair of hereditary hypergraph properties

Tim Austin, Terence Tao

math.CO · · cited 42× · arXiv ↗ · more like this ·

Recent works of Alon-Shapira and Rödl-Schacht have demonstrated that every hereditary property of undirected graphs or hypergraphs is testable with one-sided error; informally, this means that if a graph or hypergraph satisfies that property "locally" with sufficiently high probability, then it can be perturbed (or "repaired") into a graph or hypergraph which satisfies that property "globally". In this paper we make some refinements to these results, some of which may be surprising. In the positive direction, we strengthen the results to cover hereditary properties of multiple directed polychromatic graphs and hypergraphs. In the case of undirected graphs, we extend the result to continuous graphs on probability spaces, and show that the repair algorithm is "local" in the sense that it only depends on a bounded amount of data; in particular, the graph can be repaired in a time linear in…

3 Exchangeable random measures

Tim Austin

Statistics and Probability · · cited 26× · OpenAlex ↗ · more like this ·

Let A be a standard Borel space, and consider the space A^{\bbN^{(k)}} of A-valued arrays indexed by all size-k subsets of \bbN. This paper concerns random measures on such a space whose laws are invariant under the natural action of permutations of \bbN. The main result is a representation theorem for such `exchangeable' random measures, obtained using the classical representation theorems for exchangeable arrays due to de Finetti, Hoover, Aldous and Kallenberg. After proving this representation, two applications of exchangeable random measures are given. The first is a short new proof of the Dovbysh-Sudakov Representation Theorem for exchangeable PSD matrices. The second is in the formulation of a natural class of limit objects for dilute mean-field spin glass models, retaining more information than just the limiting Gram-de Finetti matrix used in the study of the…

Nonconventional ergodic averages and multiple recurrence for von Neumann dynamical systems

Tim Austin, Tanja Eisner, Terence Tao

math.OA · · cited 24× · arXiv ↗ · more like this ·

The Furstenberg recurrence theorem (or equivalently, Szemerédi's theorem) can be formulated in the language of von Neumann algebras as follows: given an integer $k \geq 2$, an abelian finite von Neumann algebra $(\M,τ)$ with an automorphism $α: \M \to \M$, and a non-negative $a \in \M$ with $τ(a)>0$, one has $\liminf_{N \to \infty} \frac{1}{N} \sum_{n=1}^N \Re τ(a α^n (a) ... α^{(k-1)n} (a)) > 0$; a subsequent result of Host and Kra shows that this limit exists. In particular, $\Re τ(a α^n (a) >... α^{(k-1)n} (a)) > 0$ for all $n$ in a set of positive density. From the von Neumann algebra perspective, it is thus natural to ask to what extent these results remain true when the abelian hypothesis is dropped. All three claims hold for $k = 2$, and we show in this paper that all three claims hold for all $k$ when the von Neumann algebra is asymptotically abelian, and that the last two…

The wreath product of Z with Z has Hilbert compression exponent 2/3

Tim Austin, Assaf Naor, Yuval Peres

math.MG · · cited 22× · arXiv ↗ · more like this ·

Let G be a finitely generated group, equipped with the word metric d associated with some finite set of generators. The Hilbert compression exponent of G is the supremum over all $α\ge 0$ such that there exists a Lipschitz mapping $f:G\to L_2$ and a constant $c>0$ such that for all $x,y\in G$ we have $\|f(x)-f(y)\|_2\ge cd(x,y)^α.$ In \cite{AGS06} it was shown that the Hilbert compression exponent of the wreath product $\Z\bwr \Z $ is at most $\frac34$, and in \cite{NP07} was proved that this exponent is at least $\frac23$. Here we show that $\frac23$ is the correct value. Our proof is based on an application of K. Ball's notion of Markov type.

The Euclidean distortion of the lamplighter group

Tim Austin, Assaf Naor, Alain Valette

math.MG · · cited 18× · arXiv ↗ · more like this ·

We show that the cyclic lamplighter group $C_2 \bwr C_n$ embeds into Hilbert space with distortion ${\rm O}(\sqrt{\log n})$. This matches the lower bound proved by Lee, Naor and Peres in \cite{LeeNaoPer}, answering a question posed in that paper. Thus the Euclidean distortion of $C_2 \bwr C_n$ is $Θ(\sqrt{\log n})$. Our embedding is constructed explicitly in terms of the irreducible representations of the group. Since the optimal Euclidean embedding of a finite group can always be chosen to be equivariant, as shown by Aharoni, Maurey and Mityagin \cite{AhaMauMit} and by Gromov (see \cite{deCTesVal}), such representation-theoretic considerations suggest a general tool for obtaining upper and lower bounds on Euclidean embeddings of finite groups.

The wreath product of $\mathbb {Z}$ with $\mathbb {Z}$ has Hilbert compression exponent $\frac {2}{3}$

Tim Austin, Assaf Naor, Yuval Peres

Mathematical Physics · Proceedings of the American Mathematical Society · · cited 16× · DOI ↗ · more like this ·

Let $G$ be a finitely generated group, equipped with the word metric $d$ associated with some finite set of generators. The Hilbert compression exponent of $G$ is the supremum over all $\alpha \ge 0$ such that there exists a Lipschitz mapping $f:G\to L_2$ and a constant $c>0$ such that for all $x,y\in G$ we have $\|f(x)-f(y)\|_2\ge cd(x,y)^\alpha .$ It was previously known that the Hilbert compression exponent of the wreath product $\mathbb {Z} {\boldsymbol {\wr }} \mathbb {Z}$ is between $\frac 23$ and $\frac 34$. Here we show that $\frac 23$ is the correct value. Our proof is based on an application of K. Ball’s notion of Markov type.

Multiple recurrence and the structure of probability-preserving systems

Tim Austin

math.DS · · cited 9× · arXiv ↗ · more like this ·

In 1975 Szemerédi proved the long-standing conjecture of Erdős and Turán that any subset of $\bbZ$ having positive upper Banach density contains arbitrarily long arithmetic progressions. Szemerédi's proof was entirely combinatorial, but two years later Furstenberg gave a quite different proof of Szemerédi's Theorem by first showing its equivalence to an ergodic-theoretic assertion of multiple recurrence, and then bringing new machinery in ergodic theory to bear on proving that. His ergodic-theoretic approach subsequently yielded several other results in extremal combinatorics, as well as revealing a range of new phenomena according to which the structures of probability-preserving systems can be described and classified. In this work I survey some recent advances in understanding these ergodic-theoretic structures. It contains proofs of the norm convergence of the `nonconventional'…

A proof of Walsh's convergence theorem using couplings

Tim Austin

math.DS · · cited 8× · arXiv ↗ · more like this ·

Walsh has recently proved the norm convergence of all nonconventional ergodic averages involving polynomial sequences in discrete nilpotent acting groups. He deduces this convergence from an equivalent, `finitary' assertion of stability over arbitrarily long time-intervals for these averages, which is proved by essentially finitary means. The present paper shows how the induction at the heart of Walsh's proof can also be implemented using more classical notions of ergodic theory: in particular, couplings and characteristic factors.

Integrable measure equivalence for groups of polynomial growth

Tim Austin, with an Appendix by Lewis Bowen

math.GR · · cited 8× · arXiv ↗ · more like this ·

Bader, Furman and Sauer have introduced the notion of integrable measure equivalence for finitely-generated groups. This is the sub-equivalence relation of measure equivalence obtained by insisting that the relevant cocycles satisfy an integrability condition. They have used it to prove new classification results for hyperbolic groups. The present work shows that groups of polynomial growth are also quite rigid under integrable measure equivalence, in that if two such groups are equivalent then they must have bi-Lipschitz asymptotic cones. This will follow by proving that the cocycles arising from an integrable measure equivalence converge, albeit in a very weak sense, to bi-Lipschitz maps of asymptotic cones.

Extensions of probability-preserving systems by measurably-varying homogeneous spaces and applications

Tim Austin

math.DS · · cited 8× · arXiv ↗ · more like this ·

We study a generalized notion of a homogeneous skew-product extension of a probability-preserving system in which the homogeneous space fibres are allowed to vary over the ergodic decomposition of the base. The construction of such extensions rests on a simple notion of `direct integral' for a `measurable family' of homogeneous spaces, which has a number of precedents in older literature. The main contribution of the present paper is the systematic development of a formalism for handling such extensions, including non-ergodic versions of the results of Mackey describing ergodic components of such extensions, of the Furstenberg-Zimmer Structure Theory and of results of Mentzen describing the structure of automorphisms of such extensions when they are relatively ergodic. We then offer applications to two structural results for actions of several commuting transformations: firstly to…

A pair of non-homeomorphic product measures on the Cantor set

Tim D. Austin

math.CA · · cited 8× · arXiv ↗ · more like this ·

For r in [0,1] let μ_r be the Bernoulli measure on the Cantor set given as the infinite power of the measure on the two-point set with weights r and 1-r. For r and s in [0,1] it is known that the measure μ_r is continuously reducible to μ_s (that is, there is a continuous map sending μ_r to μ_s) if and only if s can be written as a certain kind of polynomial in r; in this case s is said to be binomially reducible to r. In this paper we answer in the negative the following question posed by Mauldin: Is it true that the product measures μ_r and μ_s are homeomorphic if and only if each is a continuous image of the other, or, equivalently, each of the numbers r and s is binomially reducible to the other?

Non-convergence of some non-commuting double ergodic averages

Tim Austin

math.DS · · cited 7× · arXiv ↗ · more like this ·

Let $S$ and $T$ be measure-preserving transformations of a probability space $(X,{\mathcal B},μ)$. Let $f$ be a bounded measurable functions, and consider the integrals of the corresponding `double' ergodic averages: \[\frac{1}{n}\sum_{i=0}^{n-1} \int f(S^ix)f(T^ix)\ dμ(x) \qquad (n\ge 1).\] We construct examples for which these integrals do not converge as $n\to\infty$. These include examples in which $S$ and $T$ are rigid, and hence have entropy zero, answering a question of Frantzikinakis and Host. Our proof begins with a corresponding construction for orthogonal operators on a Hilbert space, and then obtains transformations of a Gaussian measure space from them.

Quantitative equidistribution for certain quadruples in quasi-random groups

Tim Austin

math.CO · · cited 7× · arXiv ↗ · more like this ·

In a recent paper (arXiv:1211.6372), Bergelson and Tao proved that if $G$ is a $D$-quasi-random group, and $x$,$g$ are drawn uniformly and independently from $G$, then the quadruple $(g,x,gx,xg)$ is roughly equidistributed in the subset of $G^4$ defined by the constraint that the last two coordinates lie in the same conjugacy class. Their proof gives only a qualitative version of this result. The present notes gives a rather more elementary proof which improves this to an explicit polynomial bound in $D^{-1}$.

The structure of low-complexity Gibbs measures on product spaces

Tim Austin

math.PR · · cited 6× · arXiv ↗ · more like this ·

Let $K_1$, $\dots$, $K_n$ be bounded, complete, separable metric spaces. Let $λ_i$ be a Borel probability measure on $K_i$ for each $i$. Let $f:\prod_i K_i \to \mathbb{R}$ be a bounded and continuous potential function, and let $$μ(d \mathbf{x})\ \propto\ e^{f(\mathbf{x})}λ_1(d x_1)\cdots λ_n(d x_n)$$ be the associated Gibbs distribution. At each point $\mathbf{x} \in \prod_i K_i$, one can define a `discrete gradient' $\nabla f(\mathbf{x},\,\cdot\,)$ by comparing the values of $f$ at all points which differ from $\mathbf{x}$ in at most one coordinate. In case $\prod_i K_i = \{-1,1\}^n \subset \mathbb{R}^n$, the discrete gradient $\nabla f(\mathbf{x},\,\cdot\,)$ is naturally identified with a vector in $\mathbb{R}^n$. This paper shows that a `low-complexity' assumption on $\nabla f$ implies that $μ$ can be approximated by a mixture of other measures, relatively few in number, and most of…

On contractive families and a fixed-point question of Stein

Tim D. Austin

math.MG · · cited 6× · arXiv ↗ · more like this ·

In this paper we disprove the following conjectured generalization of the Contraction Mapping Theorem (due to J.D. Stein Jr.): Let X be a complete metric space and let F be a finite family of self-maps of X. Suppose there is a postive constant strictly less than 1 such that, for any two points x and y of X, some member of F contracts those points by a factor of at most that constant. Then some composition of members of F has a fixed point. We also show that the above does hold for a (continuous) commuting F containing only two maps. We conjecture that it holds for commuting F of any finite size.

Additivity properties of sofic entropy and measures on model spaces

Tim Austin

math.DS · · cited 5× · arXiv ↗ · more like this ·

Sofic entropy is an invariant for probability-preserving actions of sofic groups. It was introduced a few years ago by Lewis Bowen, and shown to extend the classical Kolmogorov-Sinai entropy from the setting of amenable groups. Some parts of Kolmogorov-Sinai entropy theory generalize to sofic entropy, but in other respects this new invariant behaves less regularly. This paper explores conditions under which sofic entropy is additive for Cartesian products of systems. It is always subadditive, but the reverse inequality can fail. We define a new entropy-notion in terms of probability distributions on the spaces of good models of an action. Using this, we prove a general lower bound for the sofic entropy of a Cartesian product in terms of separate quantities for the two factor systems involved. We also prove that this lower bound is optimal in a certain sense, and use it to derive some…

A hierarchical version of the de Finetti and Aldous-Hoover representations

Tim Austin, Dmitry Panchenko

math.PR · · cited 5× · arXiv ↗ · more like this ·

We consider random arrays indexed by the leaves of an infinitary rooted tree of finite depth, with the distribution invariant under the rearrangements that preserve the tree structure. We call such arrays hierarchically exchangeable and prove that they satisfy an analogue of de Finetti's theorem. We also prove a more general result for arrays indexed by several trees, which includes a hierarchical version of the Aldous-Hoover representation.

The Geometry of Model Spaces forProbability-Preserving Actions of SoficGroups

Tim Austin

Computational Theory and Mathematics · DOAJ (DOAJ: Directory of Open Access Journals) · · cited 4× · DOI ↗ · more like this ·

Bowen’s notion of sofic entropy is a powerful invariant for classifying probability-preserving actions of sofic groups. It can be defined in terms of the covering numbers of certain metric spaces associated to such an action, the ‘model spaces’. The metric geometry of these model spaces can exhibit various interesting features, some of which provide other invariants of the action. This paper explores an approximate connectedness property of the model spaces, and uses it give a new proof that certain groups admit factors of Bernoulli shifts which are not Bernoulli. This was originally proved by Popa. Our proof covers fewer examples than his, but provides additional information about this phenomenon.