Relatore
Descrizione
The increasing complexity and volume of astrophysical data are motivating the exploration of novel computational paradigms beyond classical high-performance computing. In this context, quantum computing and quantum machine learning (QML) are emerging as potential complementary approaches for data analysis, although their practical applicability remains under active investigation.
This contribution presents an overview of recent research on the application of quantum computing techniques to astrophysical problems, with a focus on methodological developments and computational constraints. In particular, hybrid classical–quantum models will be discussed, including Quantum Convolutional Neural Networks for the classification of transient astrophysical signals, and Quantum Graph Neural Networks for the analysis of large-scale structure. Additionally, the integration of the Quantum Fourier Transform into cosmological data analysis pipelines, such as those used for Cosmic Microwave Background map processing, will be presented.
The discussion will concentrate on the key computational challenges that currently limit the performance of QML approaches in the noisy intermediate-scale quantum (NISQ) regime. These include data encoding overhead, limited qubit resources, and constraints on circuit depth, which together often dominate the overall computational cost.
| Sessione | Calcolo, Archivi e Intelligenza Artificiale |
|---|