A Hardware Accelerator for the Goemans-Williamson Algorithm
By: D. A. Herrera-Martí, E. Guthmuller, J. Fereyre
Potential Business Impact:
Makes computers solve hard problems faster.
The combinatorial problem Max-Cut has become a benchmark in the evaluation of local search heuristics for both quantum and classical optimisers. In contrast to local search, which only provides average-case performance guarantees, the convex semidefinite relaxation of Max-Cut by Goemans and Williamson, provides worst-case guarantees and is therefore suited to both the construction of benchmarks and in applications to performance-critic scenarios. We show how extended floating point precision can be incorporated in algebraic subroutines in convex optimisation, namely in indirect matrix inversion methods like Conjugate Gradient, which are used in Interior Point Methods in the case of very large problem sizes. Also, an estimate is provided of the expected acceleration of the time to solution for a hardware architecture that runs natively on extended precision. Specifically, when using indirect matrix inversion methods like Conjugate Gradient, which have lower complexity than direct methods and are therefore used in very large problems, we see that increasing the internal working precision reduces the time to solution by a factor that increases with the system size.
Similar Papers
Comparison of Hyperplane Rounding for Max-Cut and Quantum Approximate Optimization Algorithm over Certain Regular Graph Families
Quantum Physics
Finds hard math problems for quantum computers.
Reinforced Generation of Combinatorial Structures: Applications to Complexity Theory
Machine Learning (CS)
AI finds harder math problems for computers.
Characterizing QUBO Reformulations of the Max-k-Cut Problem for Quantum Computing
Quantum Physics
Finds best answers to hard problems using quantum computers.