BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
DTSTAMP:20260430T105615Z
UID:https://www.ab.mpg.de/events/44417/332793
DTSTART:20260428T083000Z
DTEND:20260428T093000Z
CLASS:PUBLIC
CREATED:20260116T144037Z
DESCRIPTION: Data collection biases pose a persistent challenge in social n
 etwork analysis\, particularly in animal social network studies where obse
 rvations are uneven\, censored\, and incomplete. These biases can lead to 
 distorted network inference and incorrect conclusions about social behavio
 ur. We present a generative Bayesian framework based on the Social Relatio
 nship Model (SRM) that jointly estimates latent social structure while exp
 licitly accounting for sampling and node-level censoring biases. Simulatio
 n experiments reflecting realistic observational scenarios show that this 
 approach reliably recovers true social connections\, even when key individ
 uals are intermittently unobserved\, and outperforms commonly used methods
  such as permutation-based tests and linear regression models. To support 
 the practical application of such models\, we also introduce Bayesian Infe
 rence (BI)\, a cross-platform Bayesian modeling software available in Pyth
 on\, R\, and Julia. BI combines an intuitive model specification syntax wi
 th the flexibility of low-level Bayesian programming and leverages GPU acc
 eleration to enable efficient inference for high-dimensional models\, achi
 eving substantial speed gains over comparable Stan implementations. Togeth
 er\, these contributions improve both the robustness and accessibility of 
 Bayesian social network analysis under realistic data collection constrain
 ts.\nSpeaker: Sebastian Sosa
LAST-MODIFIED:20260423T112034Z
LOCATION:Bückle St. 5a\, 78467 Konstanz\, Room: Seminar room MPI-AB Bückl
 estrasse + Online
ORGANIZER;CN=Max Planck Institute of Animal Behavior:mailto:bbarrett@ab.mpg
 .de
SUMMARY:Institute Seminar by Sebastian Sosa:  Bayesian Generative Network 
URL;VALUE=URI:https://www.ab.mpg.de/events/44417/332793
END:VEVENT
END:VCALENDAR