BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20220812T074334Z
LOCATION:Foyer 2nd Floor
DTSTART;TZID=Europe/Stockholm:20220628T090000
DTEND;TZID=Europe/Stockholm:20220628T110000
UID:submissions.pasc-conference.org_PASC22_sess181_pos135@linklings.com
SUMMARY:P22 - Surrogate Modeling of Laser-Plasma-Based Ion Acceleration wi
 th Invertible Neural Networks
DESCRIPTION:Poster\n\nP22 - Surrogate Modeling of Laser-Plasma-Based Ion A
 cceleration with Invertible Neural Networks\n\nMiethlinger\n\nThe interact
 ion of overdense and/or near-critical plasmas with ultra-intense laser pul
 ses presents a promising approach to enable the development of very compac
 t sources for high-energetic ions. However, current records for maximum pr
 oton energies are still below the required values for many applications, a
 nd challenges such as stability and spectral control remain unsolved to th
 is day. In particular, significant effort per experiment and a high-dimens
 ional design space renders naive sampling approaches ineffective. Furtherm
 ore, due to the strong nonlinearities of the underlying laser-plasma physi
 cs, synthetic observations by means of particle-in-cell (PIC) simulations 
 are computationally very costly, and the maximum distance between two samp
 ling points is strongly limited as well. Consequently, in order to build u
 seful surrogate models for future data generation and experimental underst
 anding and control, a combination of highly optimized simulation codes (we
  employ PIConGPU), powerful data-based methods, such as artificial neural 
 networks, and modern sampling approaches are essential. Specifically, we e
 mploy invertible neural networks for bidirectional learning of parameter a
 nd observables, and autoencoder to reduce intermediate field data to a low
 er-dimensional latent representation.
END:VEVENT
END:VCALENDAR