arXiv subjects

Richard Y. Zhang

Publications and source records attributed to Richard Y. Zhang.

At least 19 recordsLinked to original sources

arXiv2026

Sharp Restricted Isometry Thresholds for Global Minima of Rank-Restricted Matrix LASSO

We determine the sharp restricted isometry threshold for recovery at global minima of the rank-restricted matrix LASSO. For target rank $r_{\star}$, if the rank-$k$ RIP constant satisfies $δ<δ_{\mathrm{sharp}}(k/r_{\star})$, where $δ_{\mathrm{sharp}}(t)=t/(4-t)$ for $0<t<4/3$ and $δ_{\mathrm{sharp}}(t)=\sqrt{(t-1)/t}$ for $t\ge4/3$, then every global minimizer has Frobenius error $\lesssim\sqrt{r_{\star}}λ$ for all $λ\gtrsim\|\mathcal{A}^{*}(ξ)\|_{\mathrm{op}}$ and at every search rank $r\ge r_{\star}$. The constants depend only on the RIP constant and $t=k/r_{\star}$, and in particular are independent of the search rank. When the rank restriction is inactive, the result specializes to the ordinary convex matrix LASSO. We also obtain the analogous results for sparsity-restricted vector LASSO. Conversely, we show that the threshold $δ<δ_{\mathrm{sharp}}(k/r_{\star})$ cannot be improved, due to the existence of counterexamples whose global minimizers fail to recover the ground truth.

stat.ML↗

arXiv2026

Spectral Initialization and Certification for Power System Angle Estimation

Power System State Estimation (PSSE) is commonly formulated as a nonconvex weighted least-squares (WLS) problem, making global optimality difficult both to attain and to certify. Recent work has shown that, when voltage magnitudes are known to sufficient accuracy, the remaining angle estimation subproblem can be reduced to phase synchronization and solved effectively using spectral initialization and spectral certification. This paper explains why these spectral methods succeed. We prove that their behavior is governed by the measurement error, normalized against the usual observability margin from the classical literature. Below a fixed threshold, spectral initialization recovers the true voltage angles to first-order accuracy, and the WLS estimator is unique up to a global phase. Moreover, a zero-duality-gap spectral certificate verifies recovery of this unique WLS estimate. In the noiseless observable regime, spectral initialization exactly recovers the true angles and certification is exact without local refinement. Numerical experiments on standard benchmark systems support the theory and show that the spectral methods remain effective beyond the conservative regime covered by the guarantees.

math.OC↗

arXiv2026

Sharp recovery and landscape guarantees for the nonconvex matrix LASSO

Low-rank matrix recovery can be solved to statistical optimality by convex matrix optimization under the classical assumption of restricted isometry property (RIP). However, for large problems, the convex formulation is commonly replaced by a smooth rank-constrained factored nonconvex problem for which algorithmic theory typically only guarantees convergence to second-order critical points. In this paper, we develop a sharp and statistically optimal theory for second-order critical points of the factored nonconvex matrix LASSO (nuclear-norm--regularized least-squares estimator) under RIP with particular emphasis on the overparametrized regime where the search rank $r$ exceeds the ground-truth rank $r_*$. Our recovery error bounds reveal the precise role of nuclear norm regularization, interpolating between the classical convex rate and known rates for the unregularized nonconvex problem. Complementing this positive result, we give examples showing that, contrary to popular belief, rank overparametrization does not always improve the optimization landscape even under RIP. This negative result raises questions about the fundamental statistical recovery capability of rank-constrained nonconvex approaches in comparison to convex approaches which have worse computational scaling. All of our results generalize to arbitrary convex functions with nuclear-norm regularization under restricted strong convexity and smoothness. In particular, we give sharp conditions under which second-order critical points of the nonconvex problem either (1) approximately recover low-rank approximate minima of the convex problem or (2) exactly recover a low-rank global optimum if one exists.

math.OC↗

arXiv2026

Scalable Second-order Riemannian Optimization for $K$-means Clustering

Clustering is a hard discrete optimization problem. Nonconvex approaches such as low-rank semidefinite programming (SDP) have recently demonstrated promising statistical and local algorithmic guarantees for cluster recovery. Due to the combinatorial structure of the $K$-means clustering problem, current relaxation algorithms struggle to balance their constraint feasibility and objective optimality, presenting tremendous challenges in computing the second-order critical points with rigorous guarantees. In this paper, we provide a new formulation of the $K$-means problem as a smooth unconstrained optimization over a submanifold and characterize its Riemannian structures to allow it to be solved using a second-order cubic-regularized Riemannian Newton algorithm. By factorizing the $K$-means manifold into a product manifold, we show how each Newton subproblem can be solved in linear time. Our numerical experiments show that the proposed method converges significantly faster than the state-of-the-art first-order nonnegative low-rank factorization method, while achieving similarly optimal statistical accuracy.

cs.LG↗

arXiv2026

Nonnegative Low-rank Matrix Recovery Can Have Spurious Local Minima

Low-rank matrix recovery is well-known to exhibit benign nonconvexity under the restricted isometry property (RIP): every second-order critical point is globally optimal, so local methods provably recover the ground truth. Motivated by the strong empirical performance of projected gradient methods for nonnegative low-rank recovery problems, we investigate whether this benign geometry persists when the factor matrices are constrained to be elementwise nonnegative. In the simple setting of a rank-1 nonnegative ground truth, we confirm that benign nonconvexity holds in the fully-observed case with RIP constant $δ=0$. This benign nonconvexity, however, is unstable. It fails to extend to the partially-observed case with any arbitrarily small RIP constant $δ>0$, and to higher-rank ground truths $r^{\star}>1$, regardless of how much the search rank $r\ge r^{\star}$ is overparameterized. Together, these results undermine the standard stability-based explanation for the empirical success of nonconvex methods and suggest that fundamentally different tools are needed to analyze nonnegative low-rank recovery.

math.OC↗

arXiv2025

SDP-CROWN: Efficient Bound Propagation for Neural Network Verification with Tightness of Semidefinite Programming

Neural network verifiers based on linear bound propagation scale impressively to massive models but can be surprisingly loose when neuron coupling is crucial. Conversely, semidefinite programming (SDP) verifiers capture inter-neuron coupling naturally, but their cubic complexity restricts them to only small models. In this paper, we propose SDP-CROWN, a novel hybrid verification framework that combines the tightness of SDP relaxations with the scalability of bound-propagation verifiers. At the core of SDP-CROWN is a new linear bound, derived via SDP principles, that explicitly captures $\ell_{2}$-norm-based inter-neuron coupling while adding only one extra parameter per layer. This bound can be integrated seamlessly into any linear bound-propagation pipeline, preserving the inherent scalability of such methods yet significantly improving tightness. In theory, we prove that our inter-neuron bound can be up to a factor of $\sqrt{n}$ tighter than traditional per-neuron bounds. In practice, when incorporated into the state-of-the-art $α$-CROWN verifier, we observe markedly improved verification performance on large models with up to 65 thousand neurons and 2.47 million parameters, achieving tightness that approaches that of costly SDP-based methods.

cs.LG↗

arXiv2025

Sharp Global Guarantees for Nonconvex Low-rank Recovery in the Noisy Overparameterized Regime

Recent work established that rank overparameterization eliminates spurious local minima in nonconvex low-rank matrix recovery under the restricted isometry property (RIP). But this does not fully explain the practical success of overparameterization, because real algorithms can still become trapped at nonstrict saddle points (approximate second-order points with arbitrarily small negative curvature) even when all local minima are global. Moreover, the result does not accommodate for noisy measurements, but it is unclear whether such an extension is even possible, in view of the many discontinuous and unintuitive behaviors already known for the overparameterized regime. In this paper, we introduce a novel proof technique that unifies, simplifies, and strengthens two previously competing approaches -- one based on escape directions and the other based on the inexistence of counterexample -- to provide sharp global guarantees in the noisy overparameterized regime. We show, once local minima have been converted into global minima through slight overparameterization, that near-second-order points achieve the same minimax-optimal recovery bounds (up to small constant factors) as significantly more expensive convex approaches. Our results are sharp with respect to the noise level and the solution accuracy, and hold for both the symmetric parameterization $XX^{T}$, as well as the asymmetric parameterization $UV^{T}$ under a balancing regularizer; we demonstrate that the balancing regularizer is indeed necessary.

math.OC↗

arXiv2025

Preconditioned Gradient Descent for Overparameterized Nonconvex Burer--Monteiro Factorization with Global Optimality Certification

We consider using gradient descent to minimize the nonconvex function $f(X)=ϕ(XX^{T})$ over an $n\times r$ factor matrix $X$, in which $ϕ$ is an underlying smooth convex cost function defined over $n\times n$ matrices. While only a second-order stationary point $X$ can be provably found in reasonable time, if $X$ is additionally rank deficient, then its rank deficiency certifies it as being globally optimal. This way of certifying global optimality necessarily requires the search rank $r$ of the current iterate $X$ to be overparameterized with respect to the rank $r^{\star}$ of the global minimizer $X^{\star}$. Unfortunately, overparameterization significantly slows down the convergence of gradient descent, from a linear rate with $r=r^{\star}$ to a sublinear rate when $r>r^{\star}$, even when $ϕ$ is strongly convex. In this paper, we propose an inexpensive preconditioner that restores the convergence rate of gradient descent back to linear in the overparameterized case, while also making it agnostic to possible ill-conditioning in the global minimizer $X^{\star}$.

math.OC↗

arXiv2025

Preconditioned Gradient Descent for Over-Parameterized Nonconvex Matrix Factorization

In practical instances of nonconvex matrix factorization, the rank of the true solution $r^{\star}$ is often unknown, so the rank $r$ of the model can be overspecified as $r>r^{\star}$. This over-parameterized regime of matrix factorization significantly slows down the convergence of local search algorithms, from a linear rate with $r=r^{\star}$ to a sublinear rate when $r>r^{\star}$. We propose an inexpensive preconditioner for the matrix sensing variant of nonconvex matrix factorization that restores the convergence rate of gradient descent back to linear, even in the over-parameterized case, while also making it agnostic to possible ill-conditioning in the ground truth. Classical gradient descent in a neighborhood of the solution slows down due to the need for the model matrix factor to become singular. Our key result is that this singularity can be corrected by $\ell_{2}$ regularization with a specific range of values for the damping parameter. In fact, a good damping parameter can be inexpensively estimated from the current iterate. The resulting algorithm, which we call preconditioned gradient descent or PrecGD, is stable under noise, and converges linearly to an information theoretically optimal error bound. Our numerical experiments find that PrecGD works equally well in restoring the linear convergence of other variants of nonconvex matrix factorization in the over-parameterized regime.

math.OC↗

arXiv2025

Power System State Estimation by Phase Synchronization and Eigenvectors

To estimate accurate voltage phasors from inaccurate voltage magnitude and complex power measurements, the standard approach is to iteratively refine a good initial guess using the Gauss--Newton method. But the nonconvexity of the estimation makes the Gauss--Newton method sensitive to its initial guess, so human intervention is needed to detect convergence to plausible but ultimately spurious estimates. This paper makes a novel connection between the angle estimation subproblem and phase synchronization to yield two key benefits: (1) an exceptionally high quality initial guess over the angles, known as a \emph{spectral initialization}; (2) a correctness guarantee for the estimated angles, known as a \emph{global optimality certificate}. These are formulated as sparse eigenvalue-eigenvector problems, which we efficiently compute in time comparable to a few Gauss-Newton iterations. Our experiments on the complete set of Polish, PEGASE, and RTE models show, where voltage magnitudes are already reasonably accurate, that spectral initialization provides an almost-perfect single-shot estimation of $n$ angles from $2n$ moderately noisy bus power measurements (i.e. $n$ pairs of PQ measurements), whose correctness becomes guaranteed after a single Gauss--Newton iteration. For less accurate voltage magnitudes, the performance of the method degrades gracefully; even with moderate voltage magnitude errors, the estimated voltage angles remain surprisingly accurate.

math.OC↗

arXiv2025

Simple Alternating Minimization Provably Solves Complete Dictionary Learning

This paper focuses on the noiseless complete dictionary learning problem, where the goal is to represent a set of given signals as linear combinations of a small number of atoms from a learned dictionary. There are two main challenges faced by theoretical and practical studies of dictionary learning: the lack of theoretical guarantees for practically-used heuristic algorithms and their poor scalability when dealing with huge-scale datasets. Towards addressing these issues, we propose a simple and efficient algorithm that provably recovers the ground truth when applied to the nonconvex and discrete formulation of the problem in the noiseless setting. We also extend our proposed method to mini-batch and online settings where the data is huge-scale or arrives continuously over time. At the core of our proposed method lies an efficient preconditioning technique that transforms the unknown dictionary to a near-orthonormal one, for which we prove a simple alternating minimization technique converges linearly to the ground truth under minimal conditions. Our numerical experiments on synthetic and real datasets showcase the superiority of our method compared with the existing techniques.

cs.LG↗

arXiv2025

A theory on the absence of spurious solutions for nonconvex and nonsmooth optimization

We study the set of continuous functions that admit no spurious local optima (i.e. local minima that are not global minima) which we term \textit{global functions}. They satisfy various powerful properties for analyzing nonconvex and nonsmooth optimization problems. For instance, they satisfy a theorem akin to the fundamental uniform limit theorem in the analysis regarding continuous functions. Global functions are also endowed with useful properties regarding the composition of functions and change of variables. Using these new results, we show that a class of nonconvex and nonsmooth optimization problems arising in tensor decomposition applications are global functions. This is the first result concerning nonconvex methods for nonsmooth objective functions. Our result provides a theoretical guarantee for the widely-used $\ell_1$ norm to avoid outliers in nonconvex optimization.

math.OC↗

arXiv2024

Well-conditioned Primal-Dual Interior-point Method for Accurate Low-rank Semidefinite Programming

We describe how the low-rank structure in an SDP can be exploited to reduce the per-iteration cost of a convex primal-dual interior-point method down to $O(n^{3})$ time and $O(n^{2})$ memory, even at very high accuracies. A traditional difficulty is the dense Newton subproblem at each iteration, which becomes progressively ill-conditioned as progress is made towards the solution. Preconditioners have been proposed to improve conditioning, but these can be expensive to set up, and fundamentally become ineffective at high accuracies, as the preconditioner itself becomes increasingly ill-conditioned. Instead, we present a well-conditioned reformulation of the Newton subproblem that is cheap to set up, and whose condition number is guaranteed to remain bounded over all iterations of the interior-point method. In theory, applying an inner iterative method to the reformulation reduces the per-iteration cost of the outer interior-point method to $O(n^{3})$ time and $O(n^{2})$ memory. We also present a well-conditioned preconditioner that theoretically increases the outer per-iteration cost to $O(n^{3}r^{3})$ time and $O(n^{2}r^{2})$ memory, where $r$ is an upper-bound on the solution rank, but in practice greatly improves the convergence of the inner iterations.

math.OC↗

arXiv2024

Complexity of Chordal Conversion for Sparse Semidefinite Programs with Small Treewidth

If a sparse semidefinite program (SDP), specified over $n\times n$ matrices and subject to $m$ linear constraints, has an aggregate sparsity graph $G$ with small treewidth, then chordal conversion will sometimes allow an interior-point method to solve the SDP in just $O(m+n)$ time per-iteration, which is a significant speedup over the $Ω(n^{3})$ time per-iteration for a direct application of the interior-point method. Unfortunately, the speedup is not guaranteed by an $O(1)$ treewidth in $G$ that is independent of $m$ and $n$, as a diagonal SDP would have treewidth zero but can still necessitate up to $Ω(n^{3})$ time per-iteration. Instead, we construct an extended aggregate sparsity graph $\bar{G}\supseteq G$ by forcing each constraint matrix $A_{i}$ to be its own clique in $G$. We prove that a small treewidth in $\bar{G}$ does indeed guarantee that chordal conversion will solve the SDP in $O(m+n)$ time per-iteration, to $ε$-accuracy in at most $O(\sqrt{m+n}\log(1/ε))$ iterations. This sufficient condition covers many successful applications of chordal conversion, including the MAX-$k$-CUT relaxation, the Lovász theta problem, sensor network localization, polynomial optimization, and the AC optimal power flow relaxation, thus allowing theory to match practical experience.

math.OC↗

arXiv2024

Improved Global Guarantees for the Nonconvex Burer--Monteiro Factorization via Rank Overparameterization

We consider minimizing a twice-differentiable, $L$-smooth, and $μ$-strongly convex objective $ϕ$ over an $n\times n$ positive semidefinite matrix $M\succeq0$, under the assumption that the minimizer $M^{\star}$ has low rank $r^{\star}\ll n$. Following the Burer--Monteiro approach, we instead minimize the nonconvex objective $f(X)=ϕ(XX^{T})$ over a factor matrix $X$ of size $n\times r$. This substantially reduces the number of variables from $O(n^{2})$ to as few as $O(n)$ and also enforces positive semidefiniteness for free, but at the cost of giving up the convexity of the original problem. In this paper, we prove that if the search rank $r\ge r^{\star}$ is overparameterized by a \emph{constant factor} with respect to the true rank $r^{\star}$, namely as in $r>\frac{1}{4}(L/μ-1)^{2}r^{\star}$, then despite nonconvexity, local optimization is guaranteed to globally converge from any initial point to the global optimum. This significantly improves upon a previous rank overparameterization threshold of $r\ge n$, which we show is sharp in the absence of smoothness and strong convexity, but would increase the number of variables back up to $O(n^{2})$. Conversely, without rank overparameterization, we prove that such a global guarantee is possible if and only if $ϕ$ is almost perfectly conditioned, with a condition number of $L/μ<3$. Therefore, we conclude that a small amount of overparameterization can lead to large improvements in theoretical guarantees for the nonconvex Burer--Monteiro factorization.

math.OC↗

arXiv2024

Statistically Optimal K-means Clustering via Nonnegative Low-rank Semidefinite Programming

$K$-means clustering is a widely used machine learning method for identifying patterns in large datasets. Recently, semidefinite programming (SDP) relaxations have been proposed for solving the $K$-means optimization problem, which enjoy strong statistical optimality guarantees. However, the prohibitive cost of implementing an SDP solver renders these guarantees inaccessible to practical datasets. In contrast, nonnegative matrix factorization (NMF) is a simple clustering algorithm widely used by machine learning practitioners, but it lacks a solid statistical underpinning and theoretical guarantees. In this paper, we consider an NMF-like algorithm that solves a nonnegative low-rank restriction of the SDP-relaxed $K$-means formulation using a nonconvex Burer--Monteiro factorization approach. The resulting algorithm is as simple and scalable as state-of-the-art NMF algorithms while also enjoying the same strong statistical optimality guarantees as the SDP. In our experiments, we observe that our algorithm achieves significantly smaller mis-clustering errors compared to the existing state-of-the-art while maintaining scalability.

stat.ML↗

arXiv2024

Fast and Accurate Estimation of Low-Rank Matrices from Noisy Measurements via Preconditioned Non-Convex Gradient Descent

Non-convex gradient descent is a common approach for estimating a low-rank $n\times n$ ground truth matrix from noisy measurements, because it has per-iteration costs as low as $O(n)$ time, and is in theory capable of converging to a minimax optimal estimate. However, the practitioner is often constrained to just tens to hundreds of iterations, and the slow and/or inconsistent convergence of non-convex gradient descent can prevent a high-quality estimate from being obtained. Recently, the technique of preconditioning was shown to be highly effective at accelerating the local convergence of non-convex gradient descent when the measurements are noiseless. In this paper, we describe how preconditioning should be done for noisy measurements to accelerate local convergence to minimax optimality. For the symmetric matrix sensing problem, our proposed preconditioned method is guaranteed to locally converge to minimax error at a linear rate that is immune to ill-conditioning and/or over-parameterization. Using our proposed preconditioned method, we perform a 60 megapixel medical image denoising task, and observe significantly reduced noise levels compared to previous approaches.

math.OC↗

arXiv2023

Tight Certification of Adversarially Trained Neural Networks via Nonconvex Low-Rank Semidefinite Relaxations

Adversarial training is well-known to produce high-quality neural network models that are empirically robust against adversarial perturbations. Nevertheless, once a model has been adversarially trained, one often desires a certification that the model is truly robust against all future attacks. Unfortunately, when faced with adversarially trained models, all existing approaches have significant trouble making certifications that are strong enough to be practically useful. Linear programming (LP) techniques in particular face a "convex relaxation barrier" that prevent them from making high-quality certifications, even after refinement with mixed-integer linear programming (MILP) and branch-and-bound (BnB) techniques. In this paper, we propose a nonconvex certification technique, based on a low-rank restriction of a semidefinite programming (SDP) relaxation. The nonconvex relaxation makes strong certifications comparable to much more expensive SDP methods, while optimizing over dramatically fewer variables comparable to much weaker LP methods. Despite nonconvexity, we show how off-the-shelf local optimization algorithms can be used to achieve and to certify global optimality in polynomial time. Our experiments find that the nonconvex relaxation almost completely closes the gap towards exact certification of adversarially trained models.

cs.LG↗