Supreme Hornet: A Quantum Optimization Solver

 

One of the main applications of quantum computers is solving NP-hard combinatorial optimisation problems efficiently. With classical computation paradigms, in most of the cases, all we have are heuristic algorithms with which to tackle these computationally intensive problems. Therefore, it worth seeking quantum algorithms with the ability to find higher-quality solutions for such high-valued problems. A wide range of real-world optimisation problems can be recast into higher order binary optimisation (HOBO) and quadratic unconstrained binary optimisation problems (QUBOs).

In the case of HOBO, the natural prescription is to transform a higher-order problem into a quadratic one (quadratisation). Moreover, for solving very large versions of this problem, it has been shown that decomposing large QUBO instances into mini-QUBOs and merging the sub-solutions can provide an effective solution. QUBOs can be connected to Ising Hamiltonian via linear transformations. The QAOA also can be applied to the Ising Hamiltonian type problems. This will pave the way for solving QUBOs on gate-based quantum computers.

Supreme Hornet (SH) comes with a set of quantum solvers which enables users to obtain approximate solutions for large scale combinatorial optimization problems recasted in MAXCUT, MAXSAT and QUBOs.