Positive solutions for large random linear systems

DOI10.1090/PROC/15383MaRDI QIDQ4985368zbMATH OpenOpenAlexFDO

Authors Pierre Bizeul, Jamal Najim

Publication date 23 April 2021

Published in Proceedings of the American Mathematical Society (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1904.04559

zbMATH Keywords

linear systemsLotka-Volterra systems

Mathematics Subject Classification ID

Linear equations (linear algebraic aspects) (15A06) Random matrices (algebraic aspects) (15B52) Random matrices (probabilistic aspects) (60B20) Extreme value theory; extremal stochastic processes (60G70) Ecology (92D40)

Abstract: Consider a large linear system where $A_n$ is a $nimesn$ matrix with independent real standard Gaussian entries, $$ is a $nimes1$ vector of ones and with unknown the $nimes1$ vector $$ satisfying${\displaystyle }$�oldsymbol{x}_n = �oldsymbol{1}_n +frac 1{alpha_nsqrt{n}} A_n �oldsymbol{x}_n, .${\displaystyle }$We investigate the (componentwise) positivity of the solution $$ depending on the scaling factor $alph{a}_n$ as the dimension $n$ goes to $infty$. We prove that there is a sharp phase transition at the threshold $alph{a}_n^{\ast }=sqrt2logn$: below the threshold ($alph{a}_{n}llsqrt2logn$), $$ has negative components with probability tending to 1 while above ($alph{a}_{n}ggsqrt2logn$), all the vector's components are eventually positive with probability tending to 1. At the critical scaling $alph{a}_n^{\ast }$, we provide a heuristics to evaluate the probability that $$ is positive.Such linear systems arise as solutions at equilibrium of large Lotka-Volterra systems of differential equations, widely used to describe large biological communities with interactions such as foodwebs for instance. In the domaine of positivity of the solution $$, that is when $alph{a}_{n}ggsqrt2logn$, we establish that the Lotka-Volterra system of differential equations whose solution at equilibrium is precisely $$ is stable in the sense that its jacobian ${\displaystyle }${mathcal J}(�oldsymbol{x}_n) = mathrm{diag}(�oldsymbol{x}_n)left(-I_n + frac {A_n}{alpha_nsqrt{n}} ight)${\displaystyle }$ has all its eigenvalues with negative real part with probability tending to one. Our results shed a new light and complement the understanding of feasibility and stability issues for large biological communities with interaction.

Recommendations

• Positive solutions of a class of linear systems
• Positive solutions of positive linear systems
• Equilibrium in a large Lotka-Volterra system with pairwise correlated interactions
• Limiting properties of large systems of random linear equations
• scientific article; zbMATH DE number 24453

Cites work

• A chaos hypothesis for some large systems of random equations
• Concentration inequalities. A nonasymptotic theory of independence
• scientific article; zbMATH DE number 52553 (Why is no real title available?)
• scientific article; zbMATH DE number 6125590 (Why is no real title available?)
• scientific article; zbMATH DE number 3239038 (Why is no real title available?)
• Maxima of Poisson-like variables and related triangular arrays
• Spectral analysis of large dimensional random matrices
• The concentration of measure phenomenon
• The spectral radius of large random matrices

Cited in

(10)

• Positive solutions of a class of linear systems
• On nonnegative solutions of random systems of linear inequalities
• Equilibrium in a large Lotka-Volterra system with pairwise correlated interactions
• Feasibility of sparse large Lotka-Volterra ecosystems
• On non-negative solutions to large systems of random linear equations
• Equilibrium and surviving species in a large Lotka-Volterra system of differential equations
• Antagonistic interactions can stabilise fixed points in heterogeneous linear dynamical systems
• Equilibria of large random Lotka-Volterra systems with vanishing species: a mathematical approach
• Complex systems in ecology: a guided tour with large Lotka-Volterra models and random matrices
• Fragile vs robust multiple equilibria phases in generalized Lotka-Volterra model with non-reciprocal interactions

This page was built for publication: Positive solutions for large random linear systems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4985368)