*How can you use quantum neural networks (QNNs) to gain a quantum advantage on classical data?* We propose to use QNNs (and other quantum algorithms, including quantum signal processing) to process data in quantum sensors. 1/ Attempts over the past 7+ years to find near-term practical applications of quantum neural networks on classical data have faced a variety of challenges, including: 2/

*Is there a path to achieving a quantum computational advantage for machine learning with NISQ systems on problems of practical interest?* The community consensus seems to be _no_ for problems involving classical data (which I agree with). 1/ But, as has been pointed out often over the past few years, if you combine quantum sensing with quantum machine learning, then there appears to be a path to a combined *quantum sensing-computational advantage*. 2/

*How can you turn any physical system into a neural network?* Physical systems can be many orders of magnitude faster and more energy efficient for specific computing tasks than general-purpose, digital electronic processors. 1/5 However, there's a tricky question of how to harness this capability. We propose using physical systems trained using backpropagation as *physical neural networks*, and give a procedure for doing the training that is resilient against experimental imperfections and noise. 2/5

1/ This evening my first paper as a PI @CornellAEP came out as a preprint. Congratulations and deep thanks to my talented co-authors and colleagues, Hiro and Edwin, for joining me on the journey, and doing the lion's share of the work! arxiv.org/abs/1907.05483 2/ Our work tackles a long-standing challenge in quantum engineering: how to achieve all-to-all connectivity and full programmability between qubits built with superconducting circuits. I'll let our paper describe what we did, but I wanted to give the backstory on Twitter.