Responsabile: Mario Pasquato
Descrizione: Variable X-ray sources observed by the XMM-Netwon telescope and curated by the Extras project contain a wealth of astronomical information which is accessible only through machine learning approaches due to the large amount of data, which cannot be inspected manually.
One key machine learning task on this dataset is the identification of anomalies, in particular group anomalies, i.e. a subset of data points that represent a class of potentially interesting objects. Anomaly detection can be performed via traditional data mining approaches (e.g. isolation forests) or by leveraging deep learning to build a space of meaningful latent variables, e.g. through an autoencoder.
The student will determine which approach is more appropriate to our data and deploy it using tools such as scikit-learn and, for deep learning, PyTorch or Tensorflow. This thesis may potentially result in the discovery of a new class of interesting astrophysical X-ray sources; moreover, anomaly detection finds application to datasets across academia and industry, being a generally valuable skill to acquire.
Durata: 9-12 mesi
Laurea: Magistrale
Prerequisiti: Nessuno