Work

Optimization of the Sherrington-Kirkpatrick Hamiltonian

Year: 2019
Type: article
Abstract: . Let [Formula: see text] be a symmetric random matrix with independent and identically distributed (i.i.d.) Gaussian entries above the diagonal. We consider the problem of maximizing [Formula: see text] over binary vectors [Formula: see text]. In the language of statistical physics, this amounts to finding the ground state of the Sherrington–Kirkpatrick model of spin glasses. The asymptotic value of this optimization problem was characterized by Parisi via a celebrated variational principle, subsequently proved by Talagrand. We give an algorithm that, for any [Formula: see text], outputs [Formula: see text] such that [Formula: see text] is at least [Formula: see text] of the optimum value, with probability converging to one as [Formula: see text]. The algorithm’s time complexity is [Formula: see text]. We generalize it to matrices with i.i.d., but not necessarily Gaussian, entries, and obtain an algorithm that computes the MAXCUT of a dense Erdős–Renyi random graph to within a factor [Formula: see text]. As a side result, we prove that, at (low) nonzero temperature, the algorithm constructs approximate solutions of the Thouless–Anderson–Palmer equations.
Institution: Stanford University
Language: English
Open Access status: green
OpenAlex ID: W2906886359
FWCI: 14.65
Cites: 66
Cited by: 227
Related to: 10