Building scalable variational circuit training for machine learning tasks
Special Session (Research Track)
Hosted in Virtual Platform
DescriptionParameterized quantum circuits (PQC) have emerged as a quantum analogue of deep neural networks and can be trained for discriminative or generative tasks. Near-term quantum devices can support the training of circuits with moderate number of qubits (< 50) , and parameters. Understanding the resource requirements and computational overhead needed to train PQCs efficiently will aid in the development of scalable training methods. In this talk I will highlight several recent results which we have developed in pursuit of scalable methods for gradient-based training and error mitigation.