3 papers
gr-qc2026
APRIL: Auxiliary Physically-Redundant Information in Loss -- A physics-informed framework for parameter estimation with a gravitational-wave case study
Matteo Scialpi, Francesco Di Clemente, Leigh Smith +1
Physics-Informed Neural Networks (PINNs) embed the partial differential equations (PDEs) governing the system under study directly into the training of Neural Networks, ensuring so…
physics.comp-ph2025
Explainable autoencoder for neutron star dense matter parameter estimation
Francesco Di Clemente, Matteo Scialpi, MichaÅ Bejger
We present a physics-informed autoencoder designed to encode the equation of state of neutron stars into an interpretable latent space. In particular the input will be encoded in t…
hep-ph2025
Strange quark matter as dark matter: 40 years later, a reappraisal
Francesco Di Clemente, Marco Casolino, Alessandro Drago +2
Forty years ago Witten suggested that dark matter could be composed of macroscopic clusters of strange quark matter. This idea was very popular for several years, but it dropped ou…