Localization for heavy-tailed Anderson models

Omar Hurtado

arxiv: 2505.09079 · v1 · submitted 2025-05-14 · 🧮 math-ph · math.MP· math.SP

Localization for heavy-tailed Anderson models

Omar Hurtado This is my paper

Pith reviewed 2026-05-22 16:04 UTC · model grok-4.3

classification 🧮 math-ph math.MPmath.SP

keywords Anderson localizationheavy tailslarge deviation estimatesrandom matrix productsone-dimensional Schrödinger operatorsspectral localizationtransfer matrices

0 comments

The pith

Localization holds for one-dimensional Anderson models with heavy tails.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes that Anderson localization occurs in one-dimensional models even when the random potential follows a heavy-tailed distribution. It does this by using uniform large deviation estimates for products of random matrices associated with the transfer matrices of the system. These estimates ensure that the solutions to the Schrödinger equation grow or decay exponentially despite occasional large potential values. This matters for understanding wave localization in systems where the disorder can have rare but extreme fluctuations.

Core claim

By applying uniform large deviation estimates for random matrix products, the paper proves that the Anderson Hamiltonian with heavy-tailed disorder in one dimension has pure point spectrum with exponentially decaying eigenfunctions almost surely.

What carries the argument

uniform large deviation estimates for random matrix products, which bound the probability of atypical growth in the transfer matrices uniformly across heavy-tailed distributions

If this is right

The spectrum consists of pure point eigenvalues with exponentially localized eigenfunctions almost surely.
The resolvent or Green's function decays exponentially away from the diagonal.
Dynamical localization holds, preventing ballistic transport.
These results apply to distributions with infinite variance, extending beyond the usual bounded or light-tailed assumptions.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

This suggests that localization is robust to the specific tail behavior of the disorder distribution.
The method could potentially be adapted to other random operators where matrix product estimates are available.
Further work might explore the critical exponents or the localization length dependence on the tail index.

Load-bearing premise

The uniform large deviation estimates for the relevant random matrix products hold uniformly for the heavy-tailed distributions under consideration.

What would settle it

A concrete calculation or numerical experiment showing that for some heavy-tailed distribution satisfying the assumptions, there exists delocalized eigenstates or continuous spectrum.

read the original abstract

Using recent results on uniform large deviation estimates for random matrix products obtained by S. Raman and the author, we prove localization for one dimensional Anderson models with heavy tails.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper claims localization for 1D Anderson models with heavy tails by directly applying uniform large-deviation estimates from the author's prior work with Raman.

read the letter

This paper establishes localization for the one-dimensional Anderson model with heavy-tailed disorder by invoking uniform large-deviation estimates for random matrix products from the author's recent work with Raman. The extension to heavy tails is the main new element, since most existing localization proofs require lighter tails with finite moments. The note keeps the argument short and focuses on the application rather than re-deriving the estimates, which is efficient if the prior results transfer cleanly. That approach gives a targeted result useful for models with power-law disorder that appear in some condensed-matter settings. The central soft spot is whether the uniform LDE hypotheses actually hold for heavy-tailed distributions. These often lack exponential moments or the uniformity needed in some large-deviation theorems, and the abstract invokes the cited result without spelling out the verification for transfer matrices in this regime. A referee would need to see explicit confirmation that no extra conditions are required or that the heavy-tail class satisfies them. The reliance on the co-authored paper is not circular on its face, but soundness still hinges on that external check. This note is for specialists already working on rigorous Anderson localization and random matrix products. A reader familiar with the Raman-Hurtado estimates will see the value quickly and can judge the fit themselves. It deserves peer review because the claim is specific, the method is standard, and the potential gap is narrow enough for experts to resolve without rebuilding the argument from scratch.

Referee Report

1 major / 1 minor

Summary. The manuscript claims to establish Anderson localization for one-dimensional Schrödinger operators with heavy-tailed random potentials by applying uniform large-deviation estimates for products of random transfer matrices, as obtained in prior joint work with S. Raman.

Significance. If the cited uniform LDE results apply without additional restrictions, the work would extend localization to infinite-variance disorder regimes, providing a new tool for analyzing spectral properties in singular random potentials.

major comments (1)

[Main theorem and its proof (invocation of prior LDE result)] The central argument invokes the uniform LDE theorem from the cited prior work (Raman and the author) to control the growth of transfer-matrix products for heavy-tailed distributions. However, the manuscript does not explicitly verify that the power-law tails (index <2) satisfy the precise hypotheses of that theorem, including any uniformity, moment, or tail-decay conditions required for the estimates to hold uniformly over the distribution class. This verification is load-bearing for the localization claim.

minor comments (1)

[Abstract] The abstract and introduction should include a brief statement clarifying the precise range of tail indices for which the result holds.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their careful reading and for identifying the need for explicit verification of the hypotheses from our prior work. We address the major comment below.

read point-by-point responses

Referee: The central argument invokes the uniform LDE theorem from the cited prior work (Raman and the author) to control the growth of transfer-matrix products for heavy-tailed distributions. However, the manuscript does not explicitly verify that the power-law tails (index <2) satisfy the precise hypotheses of that theorem, including any uniformity, moment, or tail-decay conditions required for the estimates to hold uniformly over the distribution class. This verification is load-bearing for the localization claim.

Authors: We agree that an explicit verification is desirable for clarity and rigor. The one-dimensional Anderson models considered here use i.i.d. potentials with power-law tails of index α < 2; these distributions satisfy the tail-decay, moment, and uniformity conditions required by the uniform large-deviation theorem established in our prior joint work with S. Raman. In the revised manuscript we will insert a short subsection that recalls the precise hypotheses of that theorem and verifies that the heavy-tailed class under consideration meets them, thereby making the invocation of the LDE result fully transparent. revision: yes

Circularity Check

1 steps flagged

Localization proof depends on self-cited uniform LDE theorem from prior work by the author

specific steps

self citation load bearing [Abstract]
"Using recent results on uniform large deviation estimates for random matrix products obtained by S. Raman and the author, we prove localization for one dimensional Anderson models with heavy tails."

The sole load-bearing step of the proof is the direct application of the cited uniform LDE theorem. Because the theorem's authors include the present paper's author and the manuscript supplies no independent derivation or verification of the LDE estimates for the heavy-tailed case, the localization conclusion is justified only by that self-overlapping citation.

full rationale

The manuscript's derivation chain consists of invoking a uniform large-deviation estimate for random matrix products (from prior joint work with Raman) and then applying it to the transfer matrices of the 1D heavy-tailed Anderson model to obtain localization. No equations or steps inside the present paper derive the LDE estimates; they are imported wholesale via citation. This matches the self-citation-load-bearing pattern because the central claim reduces to the applicability of the cited result, whose hypotheses (uniformity over heavy-tailed distributions, moment conditions, etc.) are not re-verified or re-proved here. The citation is load-bearing and overlaps with the present author, yet the paper remains self-contained once the external theorem is granted; hence a moderate score rather than 8-10. No fitted-input, self-definitional, or ansatz-smuggling circularity is exhibited by the provided text.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The proof depends on background results from random matrix products and spectral theory for Schrödinger operators; no new free parameters or invented entities are introduced in the abstract.

axioms (1)

domain assumption Uniform large deviation estimates hold for the random matrix products arising from the heavy-tailed Anderson transfer matrices.
Invoked directly from the cited work by Raman and the author to control the growth of solutions.

pith-pipeline@v0.9.0 · 5525 in / 1033 out tokens · 44607 ms · 2026-05-22T16:04:26.813802+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

Theorem 2.4 ([HR25]). … P[|log∥T^E_[a,b]∥ − λ(E)| > nε] ≤ C L^{-p}
IndisputableMonolith/Foundation/AlexanderDuality.lean alexander_duality_circle_linking unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

We prove localization for one dimensional Anderson models with heavy tails.

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.