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.

Exact objectives of random linear programs and mean widths of random polyhedrons

On the optimal objective value of random linear programs

Integer Feasibility of Random Polytopes

Borell's inequality and mean width of random polytopes via discrete inequalities

Randomized Methods for Linear Constraints: Convergence Rates and Conditioning

Zero-One Laws for Random Feasibility Problems

Random Shadows of Fixed Polytopes

Exploiting the polyhedral geometry of stochastic linear bilevel programming

Linear Convergence of Randomized Primal-Dual Coordinate Method for Large-scale Linear Constrained Convex Programming

Minimum Cost Flows, MDPs, and $\ell_1$-Regression in Nearly Linear Time for Dense Instances

Speeding up Linear Programming using Randomized Linear Algebra

Piecewise Polyhedral Relaxations of Multilinear Optimization

An integer programming approach for linear programs with probabilistic constraints

Ellipsoid Fitting Up to a Constant

Restricted inverse optimal value problem on linear programming under weighted $l_1$ norm

An Efficient Parametric Linear Programming Solver and Application to Polyhedral Projection

A Linear Programming Approach to Attainable Cramer-Rao type Bounds and Randomness Condition

Linear Phase Transition in Random Linear Constraint Satisfaction Problem

On Design of Polyhedral Estimates in Linear Inverse Problems

Linear Programming, Width-1 CSPs, and Robust Satisfaction.

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