Lyapunov spectrum of random neural networks

David G. Clark

Home/Archive/arXiv

arXiv · 2610.12426

Lyapunov spectrum of random neural networks

David G. Clark

View original source ↗Read source document ↗Export metadata ↓

Abstract

The Lyapunov spectrum of a nonlinear recurrent neural network with random asymmetric couplings is calculated in the limit $N\to\infty$. The calculation is based on a finite-$N$ identity that expresses the cumulative distribution of Lyapunov exponents as a response function of the tangent-space dynamics, with the tangent trajectory selected by a minimum-norm condition rather than by an initial condition. A cavity method then determines this response function at large $N$ through a self-consistent single-site problem. This result establishes that the chaos in this network is extensive and gives access to diffeomorphism-invariant properties of the dynamics. This work was done in collaboration with the AI models GPT-6 Astra and Claude Opus 5.5.

Explore related subjects

cond-mat.dis-nn q-bio.NC

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

David G. Clark. 2026-10-08. Lyapunov spectrum of random neural networks. https://arxiv.org/abs/2610.12426

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Identifying Many-Body Localization in Realistic Dot Arrays

We investigate the feasibility of identifying many-body localization in semiconductor quantum dot arrays within current experimental capabilities. Using an extended Fermi-Hubbard model parameterized by realistic device conditions, we analyze the phase diagram through a set of observables with varying experimental accessibility. Our results demonstrate that signatures of MBL can, in principle, be accessed in present-day quantum dot platforms. Crucially, we show that the widely used imbalance fails to distinguish between interaction-induced freezing and genuine localization, highlighting the importance of carefully selecting and interpreting experimental probes. We identify observables that can reliably distinguish genuine MBL from interaction-induced effects, thereby establishing practical criteria for detecting localization in realistic quantum dot arrays.

cond-mat.dis-nn↗

arXiv2026

Short-Range Vector Spin Glasses: Parisi Hierarchies and Real-Space Ultrametricity

Replica symmetry breaking (RSB) is the organizing principle of mean-field spin-glass theory, yet whether a continuous Parisi hierarchy can arise in a genuinely short-range system in finite spatial dimension remains unresolved. Here we introduce a disordered model of $N$-component vector spins on a $D$-dimensional hypercubic lattice, with random matrix couplings on nearest-neighbor links and random tensor couplings on elementary plaquettes. In the large-$N$ limit, the model supports thermodynamically stable one-step, full, and one-full RSB phases, alongside replica-symmetric paramagnetic and ferromagnetic phases. We find that spatial locality constrains these ordered phases through two distinct infrared mechanisms. ($i$) Quenched random anisotropy destroys uniform ferromagnetism for $D\leq4$ through an Imry-Ma mechanism. ($ii$) A continuous Parisi hierarchy gives rise to replica Goldstone modes whose gaplessness is protected by a Ward identity. At leading order in $1/N$, these modes behave as a band of free spin waves, producing infrared-divergent fluctuations in $D\leq2$ and a Mermin-Wagner instability of the continuously broken phase. Remarkably, the Parisi hierarchy also acquires a direct real-space interpretation. For two pure states with mutual overlap $q$, a characteristic length $ξ(q)$ describes how far the differences between their frozen magnetization patterns remain correlated. Throughout a continuous sector of the Parisi hierarchy, $ξ(q)$ is independent of temperature and decreases monotonically with $q$. The ultrametric organization of pure states is therefore accompanied by an ultrametric hierarchy of real-space correlation lengths: states that separate near the root of the Parisi tree differ through long-wavelength frozen patterns, whereas states that separate farther from the root differ only on progressively shorter scales.

cond-mat.dis-nn↗

arXiv2026

Exact distribution of the output of a deep-layered machine

Deep learning builds complex global functions through the repeated composition of local functions. But the resulting distribution over global functions is unknown. We derive this distribution exactly for finite width and depth when the local functions are chosen at random. Surprisingly, depth first creates a sharply structured distribution, then erases this structure as probability drains into two absorbing states. This competition produces a crossover at a depth exponential in the number of inputs, beyond which additional layers mainly drive collapse. The transient structure may help explain why highly overparameterized networks can still generalize.

cond-mat.dis-nn↗