A random matrix model for random approximate $t$-designs

• URL: http://arxiv.org/abs/2210.07872v3
• Date: Tue, 16 Apr 2024 12:22:33 GMT
• Title: A random matrix model for random approximate $t$-designs
• Authors: Piotr Dulian, Adam Sawicki,
• Abstract summary: A random matrix model is proposed to describe the probability distribution of $delta(nu_mathcalS,t)$ for any $t$. We show that our model satisfies the so-called spectral gap conjecture, i.e. we prove that with the probability $1$ there is $tinmathbbZ_+$ such that $sup satisfieskinmathbbZ_+delta(k)=delta(t)$.
• Score: 1.534667887016089
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: For a Haar random set $\mathcal{S}\subset U(d)$ of quantum gates we consider the uniform measure $\nu_\mathcal{S}$ whose support is given by $\mathcal{S}$. The measure $\nu_\mathcal{S}$ can be regarded as a $\delta(\nu_\mathcal{S},t)$-approximate $t$-design, $t\in\mathbb{Z}_+$. We propose a random matrix model that aims to describe the probability distribution of $\delta(\nu_\mathcal{S},t)$ for any $t$. Our model is given by a block diagonal matrix whose blocks are independent, given by Gaussian or Ginibre ensembles, and their number, size and type is determined by $t$. We prove that, the operator norm of this matrix, $\delta({t})$, is the random variable to which $\sqrt{|\mathcal{S}|}\delta(\nu_\mathcal{S},t)$ converges in distribution when the number of elements in $\mathcal{S}$ grows to infinity. Moreover, we characterize our model giving explicit bounds on the tail probabilities $\mathbb{P}(\delta(t)>2+\epsilon)$, for any $\epsilon>0$. We also show that our model satisfies the so-called spectral gap conjecture, i.e. we prove that with the probability $1$ there is $t\in\mathbb{Z}_+$ such that $\sup_{k\in\mathbb{Z}_{+}}\delta(k)=\delta(t)$. Numerical simulations give convincing evidence that the proposed model is actually almost exact for any cardinality of $\mathcal{S}$. The heuristic explanation of this phenomenon, that we provide, leads us to conjecture that the tail probabilities $\mathbb{P}(\sqrt{\mathcal{S}}\delta(\nu_\mathcal{S},t)>2+\epsilon)$ are bounded from above by the tail probabilities $\mathbb{P}(\delta(t)>2+\epsilon)$ of our random matrix model. In particular our conjecture implies that a Haar random set $\mathcal{S}\subset U(d)$ satisfies the spectral gap conjecture with the probability $1$.

Related papers

• Relative volume of comparable pairs under semigroup majorization [0.0]
  We review recent results and conjectures in the case of emphmajorization relation. We prove new exact finite-$n$ results in the case of emphUT-majorization relation.
  arXiv  Detail & Related papers  (2024-10-30T16:48:59Z)
• The Communication Complexity of Approximating Matrix Rank [50.6867896228563]
  We show that this problem has randomized communication complexity $Omega(frac1kcdot n2log|mathbbF|)$. As an application, we obtain an $Omega(frac1kcdot n2log|mathbbF|)$ space lower bound for any streaming algorithm with $k$ passes.
  arXiv  Detail & Related papers  (2024-10-26T06:21:42Z)
• Provably learning a multi-head attention layer [55.2904547651831]
  Multi-head attention layer is one of the key components of the transformer architecture that sets it apart from traditional feed-forward models. In this work, we initiate the study of provably learning a multi-head attention layer from random examples. We prove computational lower bounds showing that in the worst case, exponential dependence on $m$ is unavoidable.
  arXiv  Detail & Related papers  (2024-02-06T15:39:09Z)
• A Unified Framework for Uniform Signal Recovery in Nonlinear Generative Compressed Sensing [68.80803866919123]
  Under nonlinear measurements, most prior results are non-uniform, i.e., they hold with high probability for a fixed $mathbfx*$ rather than for all $mathbfx*$ simultaneously. Our framework accommodates GCS with 1-bit/uniformly quantized observations and single index models as canonical examples. We also develop a concentration inequality that produces tighter bounds for product processes whose index sets have low metric entropy.
  arXiv  Detail & Related papers  (2023-09-25T17:54:19Z)
• Mean Estimation in High-Dimensional Binary Markov Gaussian Mixture Models [12.746888269949407]
  We consider a high-dimensional mean estimation problem over a binary hidden Markov model. We establish a nearly minimax optimal (up to logarithmic factors) estimation error rate, as a function of $|theta_*|,delta,d,n$.
  arXiv  Detail & Related papers  (2022-06-06T09:34:04Z)
• Matrix concentration inequalities and efficiency of random universal sets of quantum gates [0.0]
  For a random set $mathcalS subset U(d)$ of quantum gates we provide bounds on the probability that $mathcalS$ forms a $delta$-approximate $t$-design. We show that for $mathcalS$ drawn from an exact $t$-design the probability that it forms a $delta$-approximate $t$-design satisfies the inequality $mathbbPleft(delta geq x right)leq 2D_t
  arXiv  Detail & Related papers  (2022-02-10T23:44:09Z)
• Random matrices in service of ML footprint: ternary random features with no performance loss [55.30329197651178]
  We show that the eigenspectrum of $bf K$ is independent of the distribution of the i.i.d. entries of $bf w$. We propose a novel random technique, called Ternary Random Feature (TRF) The computation of the proposed random features requires no multiplication and a factor of $b$ less bits for storage compared to classical random features.
  arXiv  Detail & Related papers  (2021-10-05T09:33:49Z)
• Spectral properties of sample covariance matrices arising from random matrices with independent non identically distributed columns [50.053491972003656]
  It was previously shown that the functionals $texttr(AR(z))$, for $R(z) = (frac1nXXT- zI_p)-1$ and $Ain mathcal M_p$ deterministic, have a standard deviation of order $O(|A|_* / sqrt n)$. Here, we show that $|mathbb E[R(z)] - tilde R(z)|_F
  arXiv  Detail & Related papers  (2021-09-06T14:21:43Z)
• The planted matching problem: Sharp threshold and infinite-order phase transition [25.41713098167692]
  We study the problem of reconstructing a perfect matching $M*$ hidden in a randomly weighted $ntimes n$ bipartite graph. We show that if $sqrtd B(mathcalP,mathcalQ) ge 1+epsilon$ for an arbitrarily small constant $epsilon>0$, the reconstruction error for any estimator is shown to be bounded away from $0$.
  arXiv  Detail & Related papers  (2021-03-17T00:59:33Z)
• Near-Optimal Model Discrimination with Non-Disclosure [19.88145627448243]
  We first consider the case of a well-specified linear model with squared loss. We derive sample complexity of a similar form, even under misspecification.
  arXiv  Detail & Related papers  (2020-12-04T23:52:54Z)
• Near-Optimal SQ Lower Bounds for Agnostically Learning Halfspaces and ReLUs under Gaussian Marginals [49.60752558064027]
  We study the fundamental problems of agnostically learning halfspaces and ReLUs under Gaussian marginals. Our lower bounds provide strong evidence that current upper bounds for these tasks are essentially best possible.
  arXiv  Detail & Related papers  (2020-06-29T17:10:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.