Calculation Behavior of 2D Ising Spin Computing with Different Spin Decision Logics

We proposed "prompt decision logic", which has the advantage for accelerating the ground-state searches of Ising spin model. The Ising spin model describes the behavior of magnetic spins. It consists of each spin having four-nearest-neighbor (adjacent) spins. Escaping local minimum is a key for solving combinatorial optimization problems. Actually, many of the combinatorial optimization problems can be mapped onto the Ising spin model. In this report, we observed convergence properties of Ising spin computing with prompt decision logic, without any scheduling during energy convergence. Furthermore, the prompt decision logic showed better success probability than that of majority voting logic, as the problem size grows. Therefore, the prompt decision logic is useful for Ising spin computing and able to solve the combinatorial optimization problems.