BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//linuxsoftware.nz//NONSGML Joyous v1.4//EN
BEGIN:VEVENT
SUMMARY:labrador: A domain-optimized machine-learning tool for gravitation
 al wave inference
DTSTART:20260702T143000Z
DTEND:20260702T160000Z
DTSTAMP:20260705T094647Z
UID:c6f3fa23-082d-4bdb-b1fc-d35e21c08c71
SEQUENCE:2
CREATED:20260702T095332Z
DESCRIPTION:Fast and reliable parameter inference is becoming increasingly
  important for gravitational-wave astronomy\, especially as compact-binary
  catalogs grow and low-latency multimessenger follow-up becomes\, hopefull
 y\, more common. In this talk\, I will present labrador\, a domain-optimiz
 ed machine-learning framework for gravitational-wave inference based on ne
 ural posterior estimation.
LAST-MODIFIED:20260702T095700Z
LOCATION:DF Seminar Room (2-8.3)\, 2nd floor of Physics Building
URL:http://df.vps.tecnico.ulisboa.pt/pt/eventos/labrador-a-domain-optimize
 d-machine-learning-tool-for-gravitational-wave-inference/
X-ALT-DESC;FMTTYPE=text/html:<p data-block-key="khq5l">Fast and reliable p
 arameter inference is becoming increasingly important for gravitational-wa
 ve astronomy\, especially as compact-binary catalogs grow and low-latency 
 multimessenger follow-up becomes\, hopefully\, more common. In this talk\,
  I will present labrador\, a domain-optimized machine-learning framework f
 or gravitational-wave inference based on neural posterior estimation.</p>
END:VEVENT
END:VCALENDAR