Replication-based Inference Algorithms for Hard Computational Problems

TL;DR

This paper introduces a new class of inference algorithms that use replicated solutions to tackle NP-hard problems, demonstrating improved performance and simplicity over existing methods like parallel tempering.

Contribution

It presents a novel inference algorithm based on replication interactions and analyzes its effectiveness on the binary Ising perceptron problem.

Findings

The new algorithm outperforms parallel tempering in solution quality.

It requires less computational resources.

It is simpler to implement.

Abstract

Inference algorithms based on evolving interactions between replicated solutions are introduced and analyzed on a prototypical NP-hard problem - the capacity of the binary Ising perceptron. The efficiency of the algorithm is examined numerically against that of the parallel tempering algorithm, showing improved performance in terms of the results obtained, computing requirements and simplicity of implementation.