BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:HEP@VUB Colloquium: Karen Terveer on Information Field Theory
DTSTART:20260604T100000Z
DTEND:20260604T110000Z
DTSTAMP:20260607T203800Z
UID:indico-event-2482@indico.iihe.ac.be
DESCRIPTION:Speakers: Vital De Henau (VUB)\n\nReconstructing cosmic-ray ra
 dio data with Information Field Theory\nInformation Field Theory (IFT) is 
 a Bayesian framework for high-dimensional signal inference which integrate
 s physics-informed priors\, such as spatial correlations and physical cons
 traints\, directly into the reconstruction of continuous signals. Through 
 variational inference\, IFT approximates high-dimensional\, potentially no
 n-Gaussian posteriors to provide rigorous uncertainty quantification along
 side point estimates. In this talk\, I will present the application of IFT
  to the reconstruction of radio signals from extensive air showers. While 
 traditional simulation-matching methods currently set the benchmark for re
 construction precision\, they are often computationally intensive and typi
 cally provide only point estimates. Conversely\, data-driven machine-learn
 ing approaches offer remarkable inference speed but may require extensive 
 training datasets and can be challenging to interpret physically. IFT offe
 rs a complementary approach: it remains inherently physics-informed throug
 h its priors and requires no pre-training\, yet it uses many of the effici
 ent high-dimensional optimization techniques found in modern ML. The prese
 nted method for air shower reconstruction utilizes signal timing and fluen
 ce simultaneously\, achieving a precision of 25.4g/cm^2 in X_max and 12% i
 n radiation energy. Additionally\, it offers a three-order-of-magnitude sp
 eedup over standard techniques and benchmarks the first simultaneous recon
 struction of timing and fluence data for LOFAR.\n\nhttps://indico.iihe.ac.
 be/event/2482/
LOCATION:G/1-G.1.03 - J. Sacton (Building G)
URL:https://indico.iihe.ac.be/event/2482/
END:VEVENT
END:VCALENDAR