Anomaly detection with SOFIE - Digging the tiniest signals at the LHC

by Juan Antonio Aguilar Saavedra (IFT UAM-CSIC)




      → Direct link

      → Direct link
      Meeting ID: 629 9339 7552
      Passcode: 518909




Deep Learning approach to LHCb Calorimeter reconstruction using a Cellular Automaton Núria Valls Canudas, Xavier Vilasis Cardona, Míriam Calvo Gómez and Elisabet Golobardes Ribé EPJ Web Conf., 251 (2021) 04008 DOI:


With the boom of machine learning, anomaly detection methods are gaining traction as a tool to test at the LHC the ample variety of new physics models proposed up to date, and those still not imagined. SOFIE is a novel concept in anomaly detection, joining the best features of supervised methods with the model independence of unsupervised ones, whose performance largely improves over purely unsupervised tools .





Dr. J.A. Aguilar Saavedra Dr. Juan Antonio Aguilar Saavedra defended his PhD at Universidad de Granada in 2000. Postdoc at IST Lisboa from 2001 to 2006 and Ramón y Cajal at UGR from 2006 to 2010. Profesor Titular UGR 2010-2021. Dr. Aguilar currently works at IFT (CSIC) where he got his position as Científico Titular in 2021.

Your browser is out of date!

Update your browser to view this website correctly. Update my browser now