Complexity transitions in global algorithms for sparse linear systems over finite fields

Abstract

We study the computational complexity of a very basic problem, namely that of finding solutions to a very large set of random linear equations in a finite Galois Field modulo q. Using tools from statistical mechanics we are able to identify phase transitions in the structure of the solution space and to connect them to changes in performance of a global algorithm, namely Gaussian elimination. Crossing phase boundaries produces a dramatic increase in memory and CPU requirements necessary to the algorithms. In turn, this causes the saturation of the upper bounds for the running time. We illustrate the results on the specific problem of integer factorization, which is of central interest for deciphering messages encrypted with the RSA cryptosystem.

23 pages, 8 figures

References in corpus (3)

• Simplest random K-satisfiability problem
• Exact solutions for diluted spin glasses and optimization problems
• Phase coexistence and finite-size scaling in random combinatorial problems

Cited by in corpus (9)

• Survey propagation: an algorithm for satisfiability
• The Quantum Adiabatic Algorithm applied to random optimization problems: the quantum spin glass perspective
• On the cavity method for decimated random constraint satisfaction problems and the analysis of belief propagation guided decimation algorithms
• Bicoloring Random Hypergraphs
• Entropic barriers as a reason for hardness in both classical and quantum algorithms
• Optimization and Physics: On the satisfiability of random Boolean formulae
• Tensor networks for $p$-spin models
• Phase transitions in integer linear problems
• Typical rank of coin-toss power-law random matrices over GF(2)