Jeongho Bang, Junghee Ryu, Seokwon Yoo, Marcin Pawlowski, Jinhyoung Lee
We propose a general-purpose method of machine learning for quantum-algorithm design. The method is of using a quantum-classical hybrid simulator, where a "quantum student" is being taught by a "classical teacher." In other words, in our method, the learning system is supposed to evolve into a quantum algorithm for a given problem, assisted by classical feedback. Our method is applicable to design, in principle, every quantum oracle-based algorithm. As a case study we show that the quantum algorithms solving the Deutsch-Jozsa problem can be faithfully learned. Even more remarkable is that the learning time is proportional to the square root of the total number of parameters instead of the exponential dependance found in the classical cases.
View original:
http://arxiv.org/abs/1301.1132
No comments:
Post a Comment