Brennan Freeze ('23) presents the results of recent research at COPLAC's 2023 Midwest/West Regional Undergraduate Research, Scholarly, and Creative Activity Virtual Conference.
Improving the Performance of Quantum Computing Simulation
Brennan Freeze, Sonoma State University
Faculty Mentor: Suzanne Rivoire
Abstract
Quantum computers are exponentially faster than classical computers for certain types of computation and will be a breakthrough for machine learning practices. However, since classical computers are much more widely available, they are often used to simulate quantum computation. This task is inherently a losing battle, but various optimization techniques from classic linear algebra such as parsing of matrices, to domain-specific strategies exploiting the properties of quantum circuits can vastly expand what is feasible. These improvements vary in complexity to implement, however are necessary for rational runtimes. We demonstrate how a selection of these techniques can improve the performance of QC.py, a simple and flexible quantum simulation and visualization tool designed for education. These optimization practices in the future can possibly be used to find conjunctions of quantum and classical computing to create more efficient computing techniques.