Abstract:Given a sequence of $d \times d$ symmetric matrices $\{\mathbf{W}_i\}_{i=1}^n$, and a margin $\Delta > 0$, we investigate whether it is possible to find signs $(\epsilon_1, \dots, \epsilon_n) \in \{\pm 1\}^n$ such that the operator norm of the signed sum satisfies $\|\sum_{i=1}^n \epsilon_i \mathbf{W}_i\|_\mathrm{op} \leq \Delta$. Kunisky and Zhang (2023) recently introduced a random version of this problem, where the matrices $\{\mathbf{W}_i\}_{i=1}^n$ are drawn from the Gaussian orthogonal ensemble. This model can be seen as a random variant of the celebrated Matrix Spencer conjecture and as a matrix-valued analog of the symmetric binary perceptron in statistical physics. In this work, we establish a satisfiability transition in this problem as $n, d \to \infty$ with $n / d^2 \to \tau > 0$. Our contributions are twofold. First, we prove that the expected number of solutions with margin $\Delta=\kappa \sqrt{n}$ has a sharp threshold at a critical $\tau_1(\kappa)$: for $\tau < \tau_1(\kappa)$ the problem is typically unsatisfiable, while for $\tau > \tau_1(\kappa)$ the average number of solutions becomes exponentially large. Second, combining a second-moment method with recent results from Altschuler (2023) on margin concentration in perceptron-type problems, we identify a second threshold $\tau_2(\kappa)$, such that for $\tau>\tau_2(\kappa)$ the problem admits solutions with high probability. In particular, we establish for the first time that a system of $n = \Theta(d^2)$ Gaussian random matrices can be balanced so that the spectrum of the resulting matrix macroscopically shrinks compared to the typical semicircle law. Our proofs rely on deriving concentration inequalities and large deviation properties for the law of correlated Gaussian matrices under spectral norm constraints, extending results in the literature and offering insights of independent interest.

Capacity threshold for the Ising perceptron

Capacity Lower Bound for the Ising Perceptron.

A note on the capacity of the binary perceptron

Binary perceptrons capacity via fully lifted random duality theory

Scaling Limits for One-Dimensional Long-Range Percolation: Using the Corrector Method

Critical Window of The Symmetric Perceptron

Capacity of the treelike sign perceptrons neural networks with one hidden layer -- RDT based upper bounds

On the Atypical Solutions of the Symmetric Binary Perceptron

The Ising magnetization exponent on Z^2 is 1/15

Maximum Caliber Inference and the Stochastic Ising Model

Tractability from overparametrization: the example of the negative perceptron

Fl RDT based ultimate lowering of the negative spherical perceptron capacity

Optimal capacity of the Blume-Emery-Griffiths perceptron

A Capacity Scaling Law for Artificial Neural Networks

Capacity of the covariance perceptron

The sharp interface limit of an Ising game

The Ising Partition Function: Zeros and Deterministic Approximation

Average-case matrix discrepancy: satisfiability bounds

On the capacity functional of the infinite cluster of a Boolean model

Discrepancy Algorithms for the Binary Perceptron