QSI Seminar: Dr Maria Schuld, Xanadu, Encoding Classical Data into Quantum States for ML, 05/06/2020

This talk sheds light on different aspects of encoding classical data into quantum states for machine learning.

TITLE: Encoding Classical Data into Quantum States for Machine Learning
SPEAKER: Dr Maria Schuld
AFFILIATION: Quantum Research Group, Xanadu, Toronto, Canada | University of KwaZulu-Natal, Durban, South Africa
HOSTED BY: Dr Chris Ferrie, UTS Centre for Quantum Software and Information

ABSTRACT:
When quantum computers are used to process classical data - a setting investigated in the emerging field of quantum machine learning - the first step is to encode data into quantum states. In fact, this is the most important step: the way we encode classical data determines almost entirely the potential power of a quantum machine learning algorithm.
This talk sheds light on different aspects of this data encoding, from claims of exponential speedups to quantum feature maps and quantum kernel methods.
In particular, it will present the framework of quantum embeddings in which a data encoding can be adaptively learnt from data, while the circuit for optimal classification follows from well-known results in quantum information theory.

RELATED ARTICLES:
Quantum embeddings for machine learning: https://arxiv.org/abs/2001.03622

Centre for Quantum Software and Information: https://www.uts.edu.au/research-and-t...
Chris Ferrie: https://www.uts.edu.au/staff/christop...