Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2209.07032 · Progress of Theoretical and Experimental Physics

Real tensor eigenvalue/vector distributions of the Gaussian tensor model via a four-fermi theory

Naoki Sasakura

Published Dec 12, 2022• 7 citations
arXivHTMLPDF

Abstract

Abstract Eigenvalue distributions are important dynamic quantities in matrix models, and it is an interesting challenge to study corresponding quantities in tensor models. We study real tensor eigenvalue/vector distributions for real symmetric order-three random tensors with a Gaussian distribution as the simplest case. We first rewrite this problem as the computation of a partition function of a four-fermi theory with R replicated fermions. The partition function is exactly computed for some small-N,R cases, and is shown to precisely agree with Monte Carlo simulations. For large-N, it seems difficult to compute it exactly, and we apply an approximation using a self-consistency equation for two-point functions and obtain an analytic expression. It turns out that the real tensor eigenvalue distribution obtained by taking R = 1/2 is simply the Gaussian within this approximation. We compare the approximate expression with Monte Carlo simulations, and find that, if an extra overall factor depending on N is multiplied to the the expression, it agrees well with the Monte Carlo results. It is left for future study to improve the approximation for large-N to correctly derive the overall factor.

DOI
10.1093/ptep/ptac169

Related papers

Exact analytic expressions of real tensor eigenvalue distributions of Gaussian tensor model for small N

Naoki Sasakura

We obtain exact analytic expressions of real tensor eigenvalue/vector distributions of real symmetric order-three tensors with Gaussian distributions for N ≤ 8. This is achieved by explicitly computing the partition function of a zero-dimensional boson–fermion system with four interactions. The distributions are expressed by combinations of polynomial, exponential, and error functions as results of feasible complicated bosonic integrals that appear after fermionic integrations. By extrapolating the expressions and also using a previous result, we guess a large-N expression. The expressions are compared with Monte Carlo simulations, and precise agreement and good agreement are obtained with the exact and the large-N expressions, respectively. Understanding the feasibility of the integration is left for future study, which would provide a general-N analytic formula.

arXiv:2210.15129•Journal of Mathematical Physics•Jun 1, 2023•6 citations

Three Cases of Complex Eigenvalue/Vector Distributions of Symmetric Order-Three Random Tensors

Swastik Majumder, Naoki Sasakura

Abstract Random tensor models have applications in a variety of fields, such as quantum gravity, quantum information theory, mathematics of modern technologies, etc., and studying their statistical properties, e.g. tensor eigenvalue/vector distributions, is interesting and useful. Recently some tensor eigenvalue/vector distributions have been computed by expressing them as partition functions of 0D quantum field theories. In this paper, using this method, we compute three cases of complex eigenvalue/vector distributions of symmetric order-three random tensors, where the three cases can be characterized by the Lie-group invariances, $O(N,\mathbb {R})$, $O(N,\mathbb {C})$, and $U(N,\mathbb {C})$, respectively. Exact closed-form expressions of the distributions are obtained by computing partition functions of four-fermi theories, where the last case is of the “signed” distribution, which counts the distribution with a sign factor coming from a Hessian matrix. As an application, we compute the injective norm of the complex symmetric order-three random tensor in the large-N limit by computing the edge of the last signed distribution, obtaining agreement with an earlier numerical result in the literature.

arXiv:2408.01030•Progress of Theoretical and Experimental Physics•Aug 31, 2024•0 citations

Signed Eigenvalue/vector Distribution of Complex Order-Three Random Tensor

Naoki Sasakura

Abstract We compute the signed distribution of the eigenvalues/vectors of the complex order-three random tensor by computing a partition function of a four-fermi theory, where signs are from a Hessian determinant associated to each eigenvector. The issue of the presence of a continuous degeneracy of the eigenvectors is properly treated by a gauge fixing. The final expression is compactly represented by a generating function, which has an expansion whose powers are the dimensions of the tensor index spaces. A crosscheck is performed by Monte Carlo simulations. By taking the large-N limit we obtain a critical point where the behavior of the signed distribution qualitatively changes, and also the end of the signed distribution. The expected agreement of the end of the signed distribution with that of the genuine distribution provides a few applications, such as the largest eigenvalue, the geometric measure of entanglement, and the best rank-one approximation in the large-N limit.

arXiv:2404.03385•Progress of Theoretical and Experimental Physics•Apr 22, 2024•4 citations

The edge of random tensor eigenvalues with deviation

Nicolas Delporte, Naoki Sasakura

A bstract The largest eigenvalue of random tensors is an important feature of systems involving disorder, equivalent to the ground state energy of glassy systems or to the injective norm of quantum states. For symmetric Gaussian random tensors of order 3 and of size N , in the presence of a Gaussian noise, continuing the work [1], we compute the genuine and signed eigenvalue distributions, using field theoretic methods at large N combined with earlier rigorous results of [2]. We characterize the behaviour of the edge of the two distributions as the variance of the noise increases. We find two critical values of the variance, the first of which corresponding to the emergence of an outlier from the main part of the spectrum and the second where this outlier merges with the corresponding largest eigenvalue and they both become complex. We support our claims with Monte Carlo simulations. We believe that our results set the ground for a definition of pseudospectrum of random tensors based on Z-eigenvalues.

arXiv:2405.07731•Journal of High Energy Physics•Jan 14, 2025•3 citations

Joint Distributions of Eigenvectors of Symmetric Random Tensors

Naoki Sasakura

Abstract We compute the joint distributions of arbitrary numbers of eigenvectors of real and complex symmetric random tensors by the quantum field theoretical methods that were previously used to compute the mean distributions. We obtain the random matrix representations and the large-dimension asymptotics of the joint distributions. The latter can be expressed by a common function of tensor geometries, extending the universality found for the mean distributions to the joint distributions. Several crosschecks of our results are carried out using Monte Carlo computations.

arXiv:2605.09804•Progress of Theoretical and Experimental Physics•Jul 27, 2026•0 citations

k-Fold Gaussian Random Matrix Ensembles I: Forcing Structure into Random Matrices

Michael Brodskiy, Owen Lewis Howell

Random Matrix Theory is a powerful tool in applied mathematics. Three canonical models of random matrix distributions are the Gaussian Orthogonal, Unitary and Symplectic Ensembles. For matrix ensembles defined on k-fold tensor products of identical vector spaces we motivate natural generalizations of the Gaussian Ensemble family. We show how the k-fold invariant constraints are satisfied in both disordered spin models and systems with gauge symmetries, specifically quantum double models. We use Schur-Weyl duality to completely characterize the form of allowed Gaussian distributions satisfying k-fold invariant constraints. The eigenvalue distribution of our proposed ensembles is computed exactly using the Harish-Chandra integral method. For the 2-fold tensor product case, we show that the derived distribution couples eigenvalue spectrum to entanglement spectrum. Guided by representation theory, our work is a natural extension of the standard Gaussian random matrix ensembles.

arXiv:2405.01727•May 2, 2024•0 citations
Abstract Eigenvalue distributions are important dynamic quantities in matrix models, and it is an interesting challenge to study corresponding quantities in tensor models. We study real tensor eigenvalue/vector distributions for real symmetric order-three random tensors with a Gaussian distribution as the simplest case. We first rewrite this problem as the computation of a partition function of a four-fermi theory with R replicated fermions. The partition function is exactly computed for some small-N,R cases, and is shown to precisely agree with Monte Carlo simulations. For large-N, it seems difficult to compute it exactly, and we apply an approximation using a self-consistency equation for two-point functions and obtain an analytic expression. It turns out that the real tensor eigenvalue distribution obtained by taking R = 1/2 is simply the Gaussian within this approximation. We compare the approximate expression with Monte Carlo simulations, and find that, if an extra overall factor depending on N is multiplied to the the expression, it agrees well with the Monte Carlo results. It is left for future study to improve the approximation for large-N to correctly derive the overall factor.
We obtain exact analytic expressions of real tensor eigenvalue/vector distributions of real symmetric order-three tensors with Gaussian distributions for N ≤ 8. This is achieved by explicitly computing the partition function of a zero-dimensional boson–fermion system with four interactions. The distributions are expressed by combinations of polynomial, exponential, and error functions as results of feasible complicated bosonic integrals that appear after fermionic integrations. By extrapolating the expressions and also using a previous result, we guess a large-N expression. The expressions are compared with Monte Carlo simulations, and precise agreement and good agreement are obtained with the exact and the large-N expressions, respectively. Understanding the feasibility of the integration is left for future study, which would provide a general-N analytic formula.
Abstract Random tensor models have applications in a variety of fields, such as quantum gravity, quantum information theory, mathematics of modern technologies, etc., and studying their statistical properties, e.g. tensor eigenvalue/vector distributions, is interesting and useful. Recently some tensor eigenvalue/vector distributions have been computed by expressing them as partition functions of 0D quantum field theories. In this paper, using this method, we compute three cases of complex eigenvalue/vector distributions of symmetric order-three random tensors, where the three cases can be characterized by the Lie-group invariances, O(N,R)O(N,\mathbb {R})O(N,R), O(N,C)O(N,\mathbb {C})O(N,C), and U(N,C)U(N,\mathbb {C})U(N,C), respectively. Exact closed-form expressions of the distributions are obtained by computing partition functions of four-fermi theories, where the last case is of the “signed” distribution, which counts the distribution with a sign factor coming from a Hessian matrix. As an application, we compute the injective norm of the complex symmetric order-three random tensor in the large-N limit by computing the edge of the last signed distribution, obtaining agreement with an earlier numerical result in the literature.
Abstract We compute the signed distribution of the eigenvalues/vectors of the complex order-three random tensor by computing a partition function of a four-fermi theory, where signs are from a Hessian determinant associated to each eigenvector. The issue of the presence of a continuous degeneracy of the eigenvectors is properly treated by a gauge fixing. The final expression is compactly represented by a generating function, which has an expansion whose powers are the dimensions of the tensor index spaces. A crosscheck is performed by Monte Carlo simulations. By taking the large-N limit we obtain a critical point where the behavior of the signed distribution qualitatively changes, and also the end of the signed distribution. The expected agreement of the end of the signed distribution with that of the genuine distribution provides a few applications, such as the largest eigenvalue, the geometric measure of entanglement, and the best rank-one approximation in the large-N limit.
A bstract The largest eigenvalue of random tensors is an important feature of systems involving disorder, equivalent to the ground state energy of glassy systems or to the injective norm of quantum states. For symmetric Gaussian random tensors of order 3 and of size N , in the presence of a Gaussian noise, continuing the work [1], we compute the genuine and signed eigenvalue distributions, using field theoretic methods at large N combined with earlier rigorous results of [2]. We characterize the behaviour of the edge of the two distributions as the variance of the noise increases. We find two critical values of the variance, the first of which corresponding to the emergence of an outlier from the main part of the spectrum and the second where this outlier merges with the corresponding largest eigenvalue and they both become complex. We support our claims with Monte Carlo simulations. We believe that our results set the ground for a definition of pseudospectrum of random tensors based on Z-eigenvalues.
Abstract We compute the joint distributions of arbitrary numbers of eigenvectors of real and complex symmetric random tensors by the quantum field theoretical methods that were previously used to compute the mean distributions. We obtain the random matrix representations and the large-dimension asymptotics of the joint distributions. The latter can be expressed by a common function of tensor geometries, extending the universality found for the mean distributions to the joint distributions. Several crosschecks of our results are carried out using Monte Carlo computations.
Random Matrix Theory is a powerful tool in applied mathematics. Three canonical models of random matrix distributions are the Gaussian Orthogonal, Unitary and Symplectic Ensembles. For matrix ensembles defined on k-fold tensor products of identical vector spaces we motivate natural generalizations of the Gaussian Ensemble family. We show how the k-fold invariant constraints are satisfied in both disordered spin models and systems with gauge symmetries, specifically quantum double models. We use Schur-Weyl duality to completely characterize the form of allowed Gaussian distributions satisfying k-fold invariant constraints. The eigenvalue distribution of our proposed ensembles is computed exactly using the Harish-Chandra integral method. For the 2-fold tensor product case, we show that the derived distribution couples eigenvalue spectrum to entanglement spectrum. Guided by representation theory, our work is a natural extension of the standard Gaussian random matrix ensembles.