The Arrival of the Transformers for Advanced Analysis of Alert Streams


Guillermo Cabrera-Vives

Transformers are deep learning architectures that have shown to reach state-of-the-art performances across various domains. They were originally conceived for natural language processing, but recently the have been used for images, tabular data, and time series, among others. In this talk we will review the recent advances of transformers when applied to the characterization of alert streams, and how they have outpermorfed previous approaches. Transformers have become the new state-of-the-art and will play a key role in analizing data from the Vera C. Rubin Observatory and its Legacy Survey of Space and Time (LSST), propelling us towards new frontiers in data analysis and discovery.


This talk will be given in the Machine Learning with Rubin Observatory's data session.


Career Stage: 
Senior Researcher/Faculty