BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.4.0.1//EN
TZID:Europe/Paris
X-WR-TIMEZONE:Europe/Paris
BEGIN:VEVENT
UID:341@isdm.umontpellier.fr
DTSTART;TZID=Europe/Paris:20231109T140000
DTEND;TZID=Europe/Paris:20231109T140000
DTSTAMP:20260828T084545Z
URL:https://isdm.umontpellier.fr/events/sparse-graphical-linear-dynamical-
 systems/
SUMMARY:Sparse Graphical Linear Dynamical Systems
DESCRIPTION:Building 5\, St Priest campus\n\nMachine Learning in Montpellie
 r\, Theory &amp\; Practice\n\nTime-series datasets are central in numerous
  fields of science and engineering\, such as biomedicine\, Earth observati
 on\, and network analysis. Extensive research exists on state-space models
  (SSMs)\, which are powerful mathematical tools that allow for probabilist
 ic and interpretable learning on time series. Estimating the model paramet
 ers in SSMs is arguably one of the most complicated tasks\, and the inclus
 ion of prior knowledge is known to both ease the interpretation but also t
 o complicate the inferential tasks. In this talk\, I will introduce a nove
 l joint graphical modeling framework called DGLASSO (Dynamic Graphical Las
 so) [1]\, that bridges the static graphical Lasso model [2] and the causal
 -based graphical approach for the linear-Gaussian SSM in [3]. I will also 
 present a new inference method within the DGLASSO framework that implement
 s an efficient block alternating majorization-minimization algorithm. The 
 algorithm&#x27\;s convergence is established by departing from modern tool
 s from nonlinear analysis. Experimental validation on synthetic and real w
 eather variability data showcases the effectiveness of the proposed model 
 and inference algorithm.\n\nMachine Learning in Montpellier\, Theory &amp\
 ; Practice
ATTACH;FMTTYPE=image/jpeg:https://isdm.umontpellier.fr/wp-content/uploads/
 2026/06/ml-mtp-gC78d5.png
CATEGORIES:ML MTP
LOCATION:Saint Priest Campus\, Building 5\, \, 
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=\, ;X-APPLE-RADIUS=100;X-TI
 TLE=Saint Priest Campus\, Building 5:geo:0,0
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:STANDARD
DTSTART:20231029T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
END:VTIMEZONE
END:VCALENDAR