Quantum Machine Learning from Algorithms to Hardware

Explore quantum machine learning fundamentals and advanced concepts in this 59-minute seminar presentation by IBM Quantum research scientist Sona Najafi. Delve into three distinct domains of quantum machine learning before examining novel quantum generative/variational algorithms based on quantum many-body localized (MBL) dynamics. Learn how these algorithms successfully process various datasets, including MNIST handwritten digits, quantum many-body states, and non-local parity data. Discover theoretical proof demonstrating the superior expressive power of MBL generative models compared to classical counterparts, and understand how hidden units enhance learning capabilities. Investigate quantum neuromorphic computing concepts, focusing on the universal quantum perceptron (QP) built with interacting qubits and tunable coupling constants. Understand how adding tunable single-qubit rotations enables universal quantum computation, marking a significant advancement over classical perceptron limitations.