Boom and bust cycles due to pseudospectra of matrices with unimodular spectra

• URL: http://arxiv.org/abs/2402.19201v1
• Date: Thu, 29 Feb 2024 14:31:19 GMT
• Title: Boom and bust cycles due to pseudospectra of matrices with unimodular spectra
• Authors: Junaid Majeed Bhat, Ja\v{s} Bensa, and Marko \v{Z}nidari\v{c}
• Abstract summary: Naively, one would expect that the expectation value of such powers cannot grow as one increases the power. We demonstrate that, rather counterintuitively, a completely opposite behavior is possible.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: We discuss dynamics obtained by increasing powers of non-normal matrices that are roots of the identity, and therefore have all eigenvalues on the unit circle. Naively, one would expect that the expectation value of such powers cannot grow as one increases the power. We demonstrate that, rather counterintuitively, a completely opposite behavior is possible. In the limit of infinitely large matrices one can have an exponential growth. For finite matrices this exponential growth is a part of repeating cycles of exponential growths followed by exponential decays. The effect can occur if the spectrum is different than the pseudospectrum, with the exponential growth rate being given by the pseudospectrum. We show that this effect appears in a class of transfer matrices appearing in studies of two-dimensional non-interacting systems, for a matrix describing the Ehrenfest urn, as well as in previously observed purity dynamics in a staircase random circuit.

Related papers

• Random-Matrix-Induced Simplicity Bias in Over-parameterized Variational Quantum Circuits [72.0643009153473]
  We show that expressive variational ansatze enter a Haar-like universality class in which both observable expectation values and parameter gradients concentrate exponentially with system size.<n>As a consequence, the hypothesis class induced by such circuits collapses with high probability to a narrow family of near-constant functions.<n>We further show that this collapse is not unavoidable: tensor-structured VQCs, including tensor-network-based and tensor-hypernetwork parameterizations, lie outside the Haar-like universality class.
  arXiv  Detail & Related papers  (2026-01-05T08:04:33Z)
• On the origin of exponential operator growth in Hilbert space [0.0]
  We show that exponential operator growth in Hilbert space is governed by an off-diagonal decay of the operator matrix elements in the system eigenbasis.<n>When this decay is algebraic or slower, the growth rate saturates the universal bound.<n>This establishes a microscopic origin of operator growth which is independent of chaos, dimensionality or the presence of many-body interactions.
  arXiv  Detail & Related papers  (2025-11-04T18:24:26Z)
• Spectral Estimation with Free Decompression [47.81955761814048]
  We introduce a novel method of "free decompression" to estimate the spectrum of very large (impalpable) matrices.<n>Our method can be used to extrapolate from the empirical spectral densities of small submatrices to infer the eigenspectrum of extremely large (impalpable) matrices.
  arXiv  Detail & Related papers  (2025-06-13T17:49:25Z)
• Frustration graph formalism for qudit observables [0.0]
  We consider groups of d-outcome quantum observables with prime d represented by non-Hermitian unitary operators. We compute the generalized geometric measure of entanglement for qudit stabilizer subspaces.
  arXiv  Detail & Related papers  (2025-03-28T13:09:13Z)
• Quantum Hermitian conjugate and encoding unnormalized matrices [49.494595696663524]
  We develop the family of matrix-manipulation algorithms based on the encoding the matrix elements into the probability amplitudes of the pure superposition state of a certain quantum system. We introduce two extensions to these algorithms which allow (i) to perform Hermitian conjugation of matrices under consideration and (ii) to weaken the restriction to the absolute values of matrix elements unavoidably imposed by the normalization condition for a pure quantum state.
  arXiv  Detail & Related papers  (2025-03-27T08:49:59Z)
• Fundamental Limits of Matrix Sensing: Exact Asymptotics, Universality, and Applications [30.659400341011004]
  We provide rigorous equations characterizing the Bayes-optimal learning performance from a number of samples. We mathematically establish predictions obtained via non-rigorous methods from statistical physics.
  arXiv  Detail & Related papers  (2025-03-18T10:36:30Z)
• Single-Particle Universality of the Many-Body Spectral Form Factor [0.0]
  We consider systems of fermions evolved by non-interacting unitary circuits with correlated on-site potentials. When these potentials are drawn from the eigenvalue distribution of a circular random matrix ensemble, the spectral form factor (SFF) of the resulting circuit ensemble can be computed exactly without numerical sampling.
  arXiv  Detail & Related papers  (2024-10-09T18:00:00Z)
• Efficient conversion from fermionic Gaussian states to matrix product states [48.225436651971805]
  We propose a highly efficient algorithm that converts fermionic Gaussian states to matrix product states. It can be formulated for finite-size systems without translation invariance, but becomes particularly appealing when applied to infinite systems. The potential of our method is demonstrated by numerical calculations in two chiral spin liquids.
  arXiv  Detail & Related papers  (2024-08-02T10:15:26Z)
• Entrywise error bounds for low-rank approximations of kernel matrices [55.524284152242096]
  We derive entrywise error bounds for low-rank approximations of kernel matrices obtained using the truncated eigen-decomposition. A key technical innovation is a delocalisation result for the eigenvectors of the kernel matrix corresponding to small eigenvalues. We validate our theory with an empirical study of a collection of synthetic and real-world datasets.
  arXiv  Detail & Related papers  (2024-05-23T12:26:25Z)
• Positive Semidefinite Matrix Supermartingales [30.14855064043107]
  We explore the convergence and nonasymptotic maximal inequalities of supermartingales and backward submartingales in the space of positive semidefinite matrices. Our results lead to new concentration inequalities for either martingale dependent or exchangeable random symmetric matrices under a variety of tail conditions. These inequalities are usually expressed in the boundsner order, are sometimes valid simultaneously for all sample sizes or at an arbitrary data-dependent stopping time, and can often be tightened via an external randomization factor.
  arXiv  Detail & Related papers  (2024-01-28T04:22:43Z)
• Arbitrary relaxation rate under non-Hermitian matrix iterations [0.0]
  We study the exponential relaxation of observables propagated with a non-Hermitian transfer matrix. We show that the decay rate can be an arbitrary value between the second largest eigenvalue and the largest value in the pseudospectrum.
  arXiv  Detail & Related papers  (2023-12-22T11:18:35Z)
• Phantom relaxation rate of the average purity evolution in random circuits due to Jordan non-Hermitian skin effect and magic sums [0.0]
  Phantom relaxation is relaxation with a rate that is not given by a finite spectral gap. We explain how that can arise out of an ordinary-looking spectrum.
  arXiv  Detail & Related papers  (2023-06-13T16:10:21Z)
• Temporal Entanglement in Chaotic Quantum Circuits [62.997667081978825]
  The concept of space-evolution (or space-time duality) has emerged as a promising approach for studying quantum dynamics. We show that temporal entanglement always follows a volume law in time. This unexpected structure in the temporal entanglement spectrum might be the key to an efficient computational implementation of the space evolution.
  arXiv  Detail & Related papers  (2023-02-16T18:56:05Z)
• Re-Analyze Gauss: Bounds for Private Matrix Approximation via Dyson Brownian Motion [28.431572772564518]
  Given a symmetric matrix $M$ and a vector $lambda$, we present new bounds on the Frobenius-distance utility of the Gaussian mechanism for approximating $M$ by a matrix. Our bounds depend on both $lambda$ and the gaps in the eigenvalues of $M$, and hold whenever the top $k+1$ eigenvalues of $M$ have sufficiently large gaps.
  arXiv  Detail & Related papers  (2022-11-11T18:54:01Z)
• Curvature-informed multi-task learning for graph networks [56.155331323304]
  State-of-the-art graph neural networks attempt to predict multiple properties simultaneously. We investigate a potential explanation for this phenomenon: the curvature of each property's loss surface significantly varies, leading to inefficient learning.
  arXiv  Detail & Related papers  (2022-08-02T18:18:41Z)
• Coherent and incoherent superposition of transition matrix elements of the squeezing operator [0.0]
  We describe multiphoton transitions in the system "charged particle + electromagnetic radiation" We will show that in case of interaction with a thermal field, the semi-classical result yields an acceptable approximation only in the Rayleigh-Jeans limit.
  arXiv  Detail & Related papers  (2021-12-15T19:17:13Z)
• Generative Flows with Matrix Exponential [25.888286821451562]
  Generative flows models enjoy the properties of tractable exact likelihood and efficient sampling. We incorporate matrix exponential into generative flows. Our model achieves great performance on density estimation amongst generative flows models.
  arXiv  Detail & Related papers  (2020-07-19T11:18:47Z)

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.