Robustness Verification of Quantum Classifiers

Several important models of machine learning algorithms have been successfully generalized to the quantum world, with potential speedup training classical classifiers and applications data analytics in physics that can be implemented on near future computers. However, noise is a major obstacle practical implementation learning. In this work, we define formal framework for robustness verification analysis against noises. A robust bound derived an algorithm developed check whether or not respect data. particular, find adversarial examples during checking. Our approach Google's TensorFlow Quantum verify small disturbance noises, from surrounding environment. The effectiveness our confirmed by experimental results, including bits classification as "Hello World" example, phase recognition cluster excitation detection real world intractable physical problems, MNIST world.