Fakultät für Physik und Astronomie
STEPHEN PHILLIPS hostreviews.co.uk / UNSPLASH

Unsupervised machine learning and jet physics

Dr. Barry M. Dillon , Heidelberg University

In this talk I will discuss two recent projects: (i) latent space anomaly detection with Variational AutoEncoders (VAEs), (ii) the contrastive learning of jet representations (JetCLR). In the first part I will present results on boosted top/QCD jet anomaly detection with VAEs using a range of latent space structures. While in the second part I will discuss how we can use an unsupervised machine learning tool called Contrastive Learning of Representations (CLR) to define a representation of our jet data which is (approximately) invariant to the relevant symmetries in the problem and which is discriminative within the dataset it is trained on.

Particle and Astroparticle Theory Seminar
12 Jul 2021, 16:30
Max-Planck-Institut für Kernphysik

Add to calendar Add to calendar
Today:
There are no scheduled talks today.
Upcoming:
How clumpy was the intergalactic medium during reionization?
Frederick Davies
Thu, 4 Apr 2024, 11:00
TBD
TBD
Thu, 4 Apr 2024, 11:00
Recent research results from the HKU Late State stellar evolution team - Strange Galactic alignments and solving the sulphur anomaly in Planetary Nebulae
Quentin Parker
Fri, 5 Apr 2024, 11:00
TBD
TBD
Thu, 11 Apr 2024, 11:00
Explaining the simplicity of the cosmos
Prof. Neil Turok
Thu, 11 Apr 2024, 18:30