9-13 July 2018
Sofia, Bulgaria
Europe/Sofia timezone

Next generation generative neural networks for HEP

10 Jul 2018, 09:30
20m
Hall 3 (National Palace of Culture)

Hall 3

National Palace of Culture

presentation Track 6 – Machine learning and physics analysis Plenary

Speaker

Steven Andrew Farrell (Lawrence Berkeley National Lab. (US))

Description

Initial studies have suggested generative adversarial networks (GANs) have promise as fast simulations within HEP. These studies, while promising, have been insufficiently precise and also, like GANs in general, suffer from stability issues. We apply GANs to to generate full particle physics events (not individual physics objects), and to large weak lensing cosmology convergence maps. We explore recent developments in convergence, such as ‘optimal transport’ GANs; explore representations that match the structure of the detector better than images; and evaluate the precision and generalisability of the generated datasets.
In addition we demonstrate a framework making use of distributed computing on the Cori supercomputer at NERSC launched via interactive jupyter notebook sessions,. This allows for tackling high-resolution detector data; model selection and hyper-parameter tuning in a productive yet scalable deep learning environment.

Primary authors

Deborah BARD (LBL) Steven Andrew Farrell (Lawrence Berkeley National Lab. (US)) Wahid Bhimji (Lawrence Berkeley National Lab. (US)) Mustafa Mustafa (Lawrence Berkeley National Laboratory) Dr Zarija Lukic (LBNL)

Presentation Materials