Quantum Machine Learning from Algorithms to Hardware | Qiskit Seminar Series w/ Sona Najafi

Quantum Machine Learning from Algorithms to Hardware
Qiskit Seminar Series Episode 116 with Sona Najafi

Speaker: Sona Najafi
Host: Dr. Zlatko Minev

Abstract:

Quantum machine learning has become one of the most progressing fields of quantum technology with applications in quantum optimization, quantum chemistry as well as quantum simulation. In this talk first I will review three distinct domains of quantum machine learning. Consequently, I will introduce novel quantum generative/variational algorithms based on quantum many-body localized (MBL) dynamics and show that it is capable of learning a toy dataset consisting of patterns of MNIST handwritten digits, quantum data obtained from quantum many-body states, and non-local parity data. I will theoretically prove that the MBL generative model possesses more expressive power than classical models, and the introduction of hidden units boosts its learning power. Finally, I will discuss quantum neuromorphic computing that capitalizes on inherent system dynamics and introduce the universal quantum perceptron (QP) based on interacting qubits with tunable coupling constants. By adding tunable single-qubit rotations to the QP, I will demonstrate that a QP can realize universal quantum computation, which contrasts sharply with the limited computational complexity of a single classical perceptron.

Bio:

Sona Najafi is a research scientist at IBM quantum. Her main research interests lie at the interface of quantum many-body physics, quantum computing, and quantum machine learning with applications in quantum simulation, quantum generative/variational algorithms as well as quantum neuromorphic computing. Prior to joining IBM quantum, she was a research fellow at Harvard and Caltech. She is also the winner of the Google quantum award in 2019 and Goggle early hardware access in 2020.