IQI Weekly Seminar

Abstract: We present a quantum machine learning model inspired by convolutional neural networks (CNNs). After a brief discussion of classical CNNs, we describe the concrete circuit model/architecture for quantum CNNs and discuss how learning is performed. While quantum CNNs can be applied to classical or quantum input, we focus on the latter case. Specifically, we discuss how quantum CNNs can be used to detect quantum phase transitions and present results of numerical simulations for identifying a 1D SPT phase. Finally, we provide a theoretical explanation for the success of quantum CNNs, by relating our quantum CNN model to renormalization-group flow using the multi-scale entanglement renormalization ansatz (MERA) and quantum error correction.