Should We Use Parameterized Quantum Circuits for Machine Learning?

Explore the potential and limitations of parameterized quantum circuits (PQCs) for machine learning in this one-hour conference talk by Ryan Sweke from IBM Research, Almaden. Delve into the ongoing debate surrounding the use of PQCs in machine learning tasks and examine evidence both for and against specific PQC-based algorithms. Investigate the challenges of learning output distributions from local quantum circuits and their implications for quantum circuit born machines. Shift focus to supervised learning and analyze the extent to which popular PQC-based algorithms can be dequantized using classical kernel regression with random Fourier features. Gain insights into the current state of quantum machine learning and its potential advantages over classical methods.

Ryan Sweke - Should we use parameterized quantum circuits for machine learning? - IPAM at UCLA