BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Detecting Localised Density Anomalies in Multivariate Data - Max A
 utenrieth (StatsLab/IoA)
DTSTART:20260323T160000Z
DTEND:20260323T170000Z
UID:TALK244186@talks.cam.ac.uk
CONTACT:65128
DESCRIPTION:Detecting localized differences between two samples is a centr
 al task in scientific data analysis\, with applications ranging from signa
 l identification to regime-change detection and model validation. In this 
 talk\, I will present EagleEye\, a method for identifying local over- and 
 under-densities in multivariate feature spaces. EagleEye detects localized
  over- and under-densities by comparing the local neighbourhood structure 
 of two samples. Each point is assigned an anomaly score based on whether t
 he composition of its nearby neighbours is consistent with a binomial null
  model\, and these pointwise detections are then consolidated into interpr
 etable anomaly regions. The method also provides estimates of the backgrou
 nd level and signal purity of the detected regions. I will first illustrat
 e the method through a synthetic example with known localized over- and un
 der-densities. I will then demonstrate its application in a new-physics se
 arch at particle collider experiments in the presence of systematic backgr
 ound mismodelling\, and in a climate analysis study of localized changes i
 n spatiotemporal temperature-pattern recurrence. I will also present ongoi
 ng work applying EagleEye to searches for faint dwarf galaxy candidates in
  Gaia DR3.\n
LOCATION:Martin Ryle Seminar Room\, KICC
END:VEVENT
END:VCALENDAR