BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Random signed measures
DTSTART:20250710T151500
DTEND:20250710T161500
DTSTAMP:20260603T080009Z
UID:500bc3af5b5ede700f6539abb8957797c9896e0212d376285a2f26ec
CATEGORIES:Conferences - Seminars
DESCRIPTION:Riccardo Passeggeri\, Imperial College\nPoint processes and\, 
 more generally\, random measures are ubiquitous in statistics. However\, t
 hey can only take positive values\, which is a severe limitation in many s
 ituations. In this presentation\, we introduce random signed measures\, al
 so known as real-valued random measures\, and apply them to construct vari
 ous Bayesian non-parametric models.\nIn particular\, we provide an existen
 ce result for random signed measures\, allowing us to obtain a canonical d
 efinition for random signed measures and solve a long-standing open proble
 m.\nFurther\, we provide a representation of completely random signed meas
 ures (CRSMs)\, which extends the celebrated Kingman's representation for c
 ompletely random measures (CRMs) to the real-valued case. We then introduc
 e specific classes of random signed measures\, including the Skellam point
  process\, which plays the role of the Poisson point process in the real-v
 alued case\, and the Gaussian random measure. We use the theoretical resul
 ts to develop two Bayesian nonparametric models -- one for sentiment topic
  modeling and the other for random graphs -- and to investigate mean funct
 ion estimation in Bayesian nonparametric regression.
LOCATION:CM 1 100 https://plan.epfl.ch/?room==CM%201%20100
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR