pith. sign in

arxiv: 2601.08111 · v2 · submitted 2026-01-13 · 💻 cs.DS · cs.DM· math.CO· math.PR

Derandomizing Matrix Concentration Inequalities from Free Probability

Pith reviewed 2026-05-16 15:21 UTC · model grok-4.3

classification 💻 cs.DS cs.DMmath.COmath.PR
keywords matrix concentration inequalitiesderandomizationfree probabilitymatrix Spencer problemRamanujan graphsdeterministic algorithmspolynomial time algorithms
0
0 comments X

The pith

Deterministic polynomial-time algorithms construct outcomes satisfying sharp matrix concentration bounds from free probability.

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

The paper develops explicit deterministic procedures that find specific outcomes matching the guarantees of sharp matrix concentration inequalities previously obtained via free probability. A reader would care because these inequalities previously held only in a probabilistic sense, restricting their direct use in algorithmic constructions that require certainty. The resulting algorithms solve the matrix Spencer problem and produce near-Ramanujan graphs in polynomial time, showing that free-probability objects can guide efficient search rather than only analysis.

Core claim

We design polynomial time deterministic algorithms to construct outcomes that satisfy the guarantees of these inequalities. As direct consequences, we obtain polynomial time deterministic algorithms for the matrix Spencer problem and for constructing near-Ramanujan graphs. Our proofs show that the concepts and techniques in free probability are useful not only for mathematical analyses but also for efficient computations.

What carries the argument

Deterministic search procedures that navigate the analytic objects and bounds supplied by free probability while preserving their sharpness.

Load-bearing premise

The analytic objects and bounds from free probability admit efficient deterministic search procedures that preserve sharpness without additional assumptions on the input distributions.

What would settle it

An explicit collection of matrices for which every polynomial-time deterministic search guided by free-probability objects fails to meet the concentration bound, even though a randomized assignment succeeds with high probability.

read the original abstract

Recently, sharp matrix concentration inequalities~\cite{BBvH23,BvH24} were developed using the theory of free probability. In this work, we design polynomial time deterministic algorithms to construct outcomes that satisfy the guarantees of these inequalities. As direct consequences, we obtain polynomial time deterministic algorithms for the matrix Spencer problem~\cite{BJM23} and for constructing near-Ramanujan graphs. Our proofs show that the concepts and techniques in free probability are useful not only for mathematical analyses but also for efficient computations.

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.

Referee Report

0 major / 2 minor

Summary. The paper claims that sharp matrix concentration inequalities recently derived from free probability admit polynomial-time deterministic derandomization procedures. These procedures construct explicit outcomes satisfying the same tail bounds, yielding deterministic algorithms for the matrix Spencer problem and for constructing near-Ramanujan graphs. The proofs reuse the free-probability tail bounds directly without introducing extra logarithmic or dimension-dependent losses.

Significance. If the central claim holds, the work is significant because it supplies a concrete derandomization procedure whose per-step cost is polynomial in the input dimension and whose analysis re-uses the free-probability tail bounds without extra losses. This provides explicit algorithmic constructions for the matrix Spencer problem and near-Ramanujan graphs, demonstrating that free-probability techniques can be leveraged for efficient computation as well as analysis.

minor comments (2)
  1. [§2] §2: the description of the conditional-expectation step could include an explicit bound on the number of iterations needed to reach the target probability, to make the polynomial-time claim fully self-contained.
  2. [Figure 1] Figure 1: the caption does not state the matrix dimension or number of summands used in the plotted example, making it hard to verify that the empirical deviation matches the free-probability prediction.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their careful reading and positive summary of the manuscript. The recommendation for minor revision is noted, and we will incorporate any editorial or minor clarifications in the revised version. No specific major comments were provided in the report.

Circularity Check

0 steps flagged

No significant circularity; derivation applies external free-probability bounds via explicit derandomization

full rationale

The manuscript cites external sharp matrix concentration inequalities (BBvH23, BvH24) derived from free probability and supplies a concrete polynomial-time deterministic construction (via conditional expectations or greedy search) whose analysis directly re-uses those tail bounds without extra losses or self-referential fitting. No step reduces a claimed prediction to a fitted parameter by construction, no uniqueness theorem is imported from the authors' own prior work, and no ansatz is smuggled via self-citation. The central algorithmic claim therefore rests on independent prior results and remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review yields no explicit free parameters, axioms, or invented entities; the central claim rests on the existence of the cited inequalities and the feasibility of their derandomization.

pith-pipeline@v0.9.0 · 5376 in / 904 out tokens · 43831 ms · 2026-05-16T15:21:53.841987+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

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

  • Cost.FunctionalEquation washburn_uniqueness_aczel unclear
    ?
    unclear

    Relation between the paper passage and the cited Recognition theorem.

    the moments of the free model X_free can be computed efficiently in polynomial time via a natural recursive formula; see Section 6... non-crossing structure in free probability gives recursive formulas for computing moments

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.