À la Une

Autoencoding Anomalies & Understanding Uncertainties


Le prof. Gregor Kasieczka présentera le prochain séminaire du CUI intitulé:

Autoencoding Anomalies & Understanding Uncertainties

Date: Jeudi 7 novembre 2019 à 11h00

Lieu: CUI / Battelle bât. A, Auditoire rez-de-chaussée



Many experimental results from both particle and astrophysics hint that the Standard Model (SM) of particle physics cannot be a complete theory of Nature. However, in its first years of operation, the LHC was very successful in excluding large regions of parameter space for potential models beyond the SM. We present how deep learning can be used to search for deviations from the SM in a model independent way by using deep autoencoders. Beyond autoencoders, we discuss methods of improving the robustness of network decisions and quantifying uncertainties.

Short Bio:

Diploma of Physics from Vienna Technical University, PhD from Heidelberg University and Postdoc at ETH Zurich. Since 2017 Junior Professur and Emmy Noether Group leader at Hamburg University - working on exotic searches for new physics and developments of machine learning in physics.