Quantum and classical processes combine to solve mazes

Mark S. Hillery, Daniel Koch

Yuto Ishikawa, Takuma Yoshihara, Keita Okamura +1 more

In this study, the maze generation using quantum annealing is proposed. We reformulate a standard algorithm to generate a maze into a specific form of a quadratic unconstrained binary optimization problem suitable for the input of the quantum annealer. To generate more difficulty mazes, we introduce an additional cost function Qupdate to increase the difficulty. The difficulty of the mazes was evaluated by the time to solve the maze of 12 human subjects. To check the efficiency of our scheme to create the maze, we investigated the time-to-solution of a quantum processing unit, classical computer, and hybrid solver. The results show that Qupdate generates difficult mazes tailored to the individual. Furthermore, it show that the quantum processing unit is more efficient at generating mazes than other solvers. Finally, we also present applications how our results could be used in the future.

Tadashi Ogitsu

Debabrata Goswami

Many large image processing and data processing scenarios can soon develop into maze problems requiring, say, finding the longest possible path, which are unresponsive, intractable, and challenging to analyze! Such maze problems, of which the traveling salesman decision problem is a special case, are of the Non-determinate Polynomial (NPcomplete) class of problems that are often impossible to solve with finite time and storage. We propose a novel methodology to approach this class of NP-complete problems. We convert a suitably formulated maze problem into a Quantum Search Problem (QSP), and the desired solutions are then sought using the iterative Grover’s Search Algorithm. Thus, we reformulate the entire class of such NP-problems into QSPs. Our current solution deals with two-dimensional perfect mazes with no closed loops. We encode all possible individual paths from the starting point of the maze into a quantum register. A quantum fitness operator applied to the register encodes each qubit with its fitness value. We propose the design of an optical oracle that marks all entities above a certain fitness value and uses the Grover search algorithm to find the optimal marked state in an iterative manner.

Mark A Bashuk

It is well known, and appreciated, that quantum computers have the potential to be the most powerful computational devices ever created. This newfound power comes from a quantum parallelism effect that allows the computer to be in multiple states at the same time. This property of quantum parallelism, while suited to handle common problems such as factoring and searching an unorganized database, is extremely well-suited to handle the task of solving a binary maze. I propose an algorithm that can be used to solve a binary maze on a quantum computer, with guaranteed accuracy. While it does work, it does come with a few setbacks, in that the maze must have no flaws, and that the computer requires a number of qubits equal to the number of decisions in the maze, plus log 2 of the decisions.

배준우