Deep neural networks for real-time detection and characterization of gravitational waves from compact binaries

Deep neural networks are computational models with the ability to learn from observational data and have already had spectacular success in tasks such as computer vision and natural language processing. I present a deep learning framework for real-time detection, classification and parameter estimation of gravitational waves from compact binaries, with a particular attention to systems involving neutron stars. The implications for detection and interpretation of recent and future gravitational-wave signals from neutron-star binaries, and the equation of state (EOS) of dense matter will be discussed.