New perspectives on quantum matter: bringing together quantum simulations and machine learning
This talk will review applications of quantum simulators that make use of machine learning techniques. Snapshots of many-body states obtained from quantum gas microscopes can be used to perform hypothesis testing using convolutional neural networks. The application of this technique to the Fermi Hubbard model has demonstrated that geometrical string model provides a better description of the experimental data than the pi-flux RVB model. I will also discuss the idea of combing quantum simulators with machine learning to perform inference of NMR spectra for small biological molecules.
Host: Jay Sau