Abstract:SIAM Journal on Computing, Volume 53, Issue 1, Page 1-46, February 2024. We consider the problem of finding nearly optimal solutions of optimization problems with random objective functions. Such problems arise widely in the theory of random graphs, theoretical computer science, and statistical physics. Two concrete problems we consider are (a) optimizing the Hamiltonian of a spherical or Ising [math]-spin glass model and (b) finding a large independent set in a sparse Erdős–Rényi graph. The following families of algorithms are considered: (a) low-degree polynomials of the input—a general framework that captures many prior algorithms; (b) low-depth Boolean circuits; (c) the Langevin dynamics algorithm, a canonical Monte Carlo analogue of the gradient descent algorithm. We show that these families of algorithms cannot have high success probability. For the case of Boolean circuits, our results improve the state-of-the-art bounds known in circuit complexity theory (although we consider the search problem as opposed to the decision problem). Our proof uses the fact that these models are known to exhibit a variant of the overlap gap property (OGP) of near-optimal solutions. Specifically, for both models, every two solutions whose objectives are above a certain threshold are either close to or far from each other. The crux of our proof is that the classes of algorithms we consider exhibit a form of stability (noise-insensitivity): a small perturbation of the input induces a small perturbation of the output. We show by an interpolation argument that stable algorithms cannot overcome the OGP barrier. The stability of Langevin dynamics is an immediate consequence of the well-posedness of stochastic differential equations. The stability of low-degree polynomials and Boolean circuits is established using tools from Gaussian and Boolean analysis—namely hypercontractivity and total influence, as well as a novel lower bound for random walks avoiding certain subsets, which we expect to be of independent interest. In the case of Boolean circuits, the result also makes use of Linial–Mansour–Nisan's classical theorem. Our techniques apply more broadly to low influence functions, and we expect that they may apply more generally.

A note on Probably Certifiably Correct algorithms

Computing Optimality Certificates for Convex Mixed-Integer Nonlinear Problems

Certification of the proximal gradient method under fixed-point arithmetic for box-constrained QP problems

Non-Convex Optimization with Certificates and Fast Rates Through Kernel Sums of Squares

Algorithms that satisfy a stopping criterion, probably

Probably certifiably correct k-means clustering

Analyzing the numerical correctness of branch-and-bound decisions for mixed-integer programming

Certified Multifidelity Zeroth-Order Optimization

A note on solving nonlinear optimization problems in variable precision

Stability and Recovery for Independence Systems

Certified Multi-Fidelity Zeroth-Order Optimization

Exact Algorithms for Maximum Clique: a computational study

Hardness of Random Optimization Problems for Boolean Circuits, Low-Degree Polynomials, and Langevin Dynamics

On the asymptotic behavior of the Douglas-Rachford and proximal-point algorithms for convex optimization

Algorithmic Foundations of Inexact Computing

Toward Practical Benchmarks of Ising Machines: A Case Study on the Quadratic Knapsack Problem

Algorithm-independent bounds on complex optimization through the statistics of marginal optima

Optimally Stopped Optimization

An optimal method for stochastic composite optimization

Bounding Optimality Gaps for Non-Convex Optimization Problems: Applications to Nonlinear Safety-Critical Systems

Optimal Stochastic Approximation Algorithms for Strongly Convex Stochastic Composite Optimization, II: Shrinking Procedures and Optimal Algorithms