Deep learning for quantum sciences: Selected topics

Cambridge University Press eBooks, pp. 184–220

Abstract

This chapter discusses more specialized examples on how machine learning can be used to solve problems in quantum sciences. We start by explaining the concept of differentiable programming and its use cases in quantum sciences. Next, we describe deep generative models, which have proven to be an extremely appealing tool for sampling from unknown target distributions in domains ranging from high-energy physics to quantum chemistry. Finally, we describe selected machine learning applications for experimental setups such as ultracold systems or quantum dots. In particular, we show how machine learning can help in tedious and repetitive experimental tasks in quantum devices or in validating quantum simulators with Hamiltonian learning.

Authors 2

1. Borja Requena

   Institute of Photonic Sciences

   Affiliation as printed

   ICFO - The Institute of Photonic Sciences

2. Marylou Gabrié

   Institut Polytechnique de Paris

   Affiliation as printed

   Institut Polytechnique de Paris