Quantum machine learning with quantum sensors and beyond

While the emergence of quantum sensor networks has opened up the opportunity to enhance complex sensing tasks, it has also brought tremendous challenges in designing and analyzing quantum sensing protocols due to complex structure of entanglement and physical processes. Supervised learning assisted by an entangled sensor network (SLAEN) [Phys. Rev. X 9, 041023 (2019)] is a general paradigm to automate the sensor-network design via variational quantum machine learning. The first part of the talk will present recent results in extending the entangled sensor network in general quantum data classification tasks. While capability of sensors is enhanced via adding nonlinearity, the training of such networks become harder. The second part of the talk addresses some results in the trainability of such quantum variational systems [arXiv:2311.18144]. Besides data classification, entangled sensor networks generate the quantum data that allows much more information processing tasks, including generative learning of quantum data. The third part of the talk will cover some results in quantum denoising diffusion probabilistic model (QuDDPM) to enable efficiently trainable generative learning of quantum data [arXiv:2310.05866, to appear in Phys. Rev. Lett.].