BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Daniel Conde Villatoro:  Interpretability in machine learning mode
 ls using symbolic regression: Angular coefficients in Z production
DTSTART;VALUE=DATE-TIME:20260428T133000Z
DTEND;VALUE=DATE-TIME:20260428T153000Z
DTSTAMP;VALUE=DATE-TIME:20260429T023721Z
UID:indico-event-8582@indico.ific.uv.es
DESCRIPTION:Computing angular coefficients for W and Z bosons at the LHC i
 s computationally expensive and often yields results that are unwieldy to 
 use analytically. In this talk\, I show how symbolic regression offers a p
 ractical alternative: lightweight analytical formulas derived directly fro
 m NLO simulation data. Using the PySR package\, we recover closed-form exp
 ressions for the full set of angular coefficients as functions of transver
 se momentum\, rapidity\, and invariant mass. Beyond efficiency\, these exp
 ressions shed light on the underlying physics in a way that neural network
 s cannot. Our results demonstrate that symbolic regression can produce acc
 urate and generalisable expressions that match Monte Carlo predictions wit
 hin uncertainties\, while preserving interpretability and providing insigh
 t into the kinematic dependence of angular observables.\n\nhttps://indico.
 ific.uv.es/event/8582/
LOCATION:Universe 1001-Primera-1-1-1 - Paterna. Seminario
URL:https://indico.ific.uv.es/event/8582/
END:VEVENT
END:VCALENDAR