Superconcentration, and randomized Dvoretzky's theorem for spaces with 1-unconditional bases

  title={Superconcentration, and randomized Dvoretzky's theorem for spaces with 1-unconditional bases},
  author={Konstantin E. Tikhomirov},
  journal={Journal of Functional Analysis},
Abstract Let n be a sufficiently large natural number and let B be an origin-symmetric convex body in R n in the l-position, and such that the space ( R n , ‖ ⋅ ‖ B ) admits a 1-unconditional basis. Then for any e ∈ ( 0 , 1 / 2 ] , and for random c e log ⁡ n / log ⁡ 1 e -dimensional subspace E distributed according to the rotation-invariant (Haar) measure, the section B ∩ E is ( 1 + e ) -Euclidean with probability close to one. This shows that the “worst-case” dependence on e in the randomized… 
Dichotomies, structure, and concentration in normed spaces
Abstract We use probabilistic, topological and combinatorial methods to establish the following deviation inequality: For any normed space X = ( R n , ‖ ⋅ ‖ ) there exists an invertible linear map T
  • 2020
Let Cn = {−1, 1} be the discrete hypercube equipped with the uniform probability measure σn. Talagrand’s influence inequality (1994), also known as the L1 − L2 inequality, asserts that there exists C
Hypercontractivity, and Lower Deviation Estimates in Normed Spaces
We consider the problem of estimating probabilities of lower deviation $\mathbb P\{\|G\| \leqslant \delta \mathbb E\|G\|\}$ in normed spaces with respect to the Gaussian measure. These estimates
Variance estimates and almost Euclidean structure
Abstract We introduce and initiate the study of new parameters associated with any norm and any log-concave measure on ℝn, which provide sharp distributional inequalities. In the Gaussian context
On the tightness of Gaussian concentration for convex functions
  • P. Valettas
  • Mathematics
    Journal d'Analyse Mathématique
  • 2019
The concentration of measure phenomenon in Gauss' space states that every $L$-Lipschitz map $f$ on $\mathbb R^n$ satisfies \[ \gamma_{n} \left(\{ x : | f(x) - M_{f} | \geqslant t \} \right) \leqslant
The variance of the $\ell _p^n$-norm of the Gaussian vector, and Dvoretzky’s theorem
Let $n$ be a large integer, and let $G$ be the standard Gaussian vector in $R^n$. Paouris, Valettas and Zinn (2015) showed that for all $p\in[1,c\log n]$, the variance of the $\ell_p^n$--norm of $G$
Talagrand's influence inequality revisited
Let $\mathscr{C}_n=\{-1,1\}^n$ be the discrete hypercube equipped with the uniform probability measure $\sigma_n$. Talagrand's influence inequality (1994) asserts that there exists $C\in(0,\infty)$
Quelques inégalités de superconcentration : théorie et applications
Cette these porte sur le phenomene de superconcentration qui apparait dans l'etude des fluctuations de divers modeles de la recherche actuelle (matrices aleatoires, verres de spins, champ libre


Random version of Dvoretzky’s theorem in ℓpn
We study the dependence on e in the critical dimension k(n,p,e) for which one can find random sections of the lpn-ball which are (1+e)-spherical. We give lower (and upper) estimates for k(n,p,e) for
Embedding ofl∞k in finite dimensional Banach spaces
AbstractLetx1,x2, ...,xn ben unit vectors in a normed spaceX and defineMn=Ave{‖Σi=1nε1xi‖:ε1=±1}. We prove that there exists a setA⊂{1, ...,n} of cardinality $$\left| A \right| \geqq \left[ {\sqrt n
The Random Version of Dvoretzky's Theorem in l_{\infty}^n
We show that with “high probability” a section of the `∞ ball of dimension k ≤ cε log n (c > 0 a universal constant) is ε close to a multiple of the Euclidean ball in this section. We also show that,
Euclidean Sections of Convex Bodies
This is a somewhat expanded form of a 4h course given, with small variations, first at the educational workshop Probabilistic methods in geometry, Bedlewo, Poland, July 6–12, 2008 and a few weeks
On Dvoretzky's theorem for subspaces of L
We prove that for any $2 \varepsilon \mathbb E\|Z\| \right) \leq C \exp \left (- c \min \left\{ \alpha_p \varepsilon^2 n, (\varepsilon n)^{2/p} \right\} \right), \quad 0 0$ is a constant depending
On Russo's Approximate Zero-One Law
Consider the product measure μ p on {0, 1} n , when 0 (resp. 1) is given weight 1-p (resp. p). Consider a monotone subset A of {0, 1} n . We give a precise quantitative form to the following
Superconcentration and Related Topics
Preface.- 1.Introduction.- 2.Markov semigroups.- 3.Super concentration and chaos.- 4.Multiple valleys.- 5.Talagrand's method for proving super concentration.- 6.The spectral method for proving super
Two observations regarding embedding subsets of Euclidean spaces in normed spaces
Abstract This paper contains two results concerning linear embeddings of subsets of Euclidean space in low-dimensional normed spaces. The first is an improvement of the known dependence on ɛ in
Asymptotic Theory Of Finite Dimensional Normed Spaces
The Concentration of Measure Phenomenon in the Theory of Normed Spaces.- Preliminaries.- The Isoperimetric Inequality on Sn?1 and Some Consequences.- Finite Dimensional Normed Spaces, Preliminaries.-
Extremal problems and isotropic positions of convex bodies
LetK be a convex body in ℝn and letWi(K),i=1, …,n−1 be its quermassintegrals. We study minimization problems of the form min{Wi(TK)|T ∈ SLn} and show that bodies which appear as solutions of such