BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20250312T164949EDT-0842gr90R0@132.216.98.100
DTSTAMP:20250312T204949Z
DESCRIPTION:Time-Varying Mixtures of Markov Chains: An Application to Traff
 ic Modeling.\n\nTime-varying mixture models are useful for representing co
 mplex\, dynamic distributions. Components in the mixture model can appear 
 and disappear\, and persisting components can evolve. This allows great fl
 exibility in streaming data applications where the model can be adjusted a
 s new data arrives. Fitting a mixture model\, especially when the model or
 der varies with time\, with computational guarantees which can meet real-t
 ime requirements is difficult with existing algorithms. Multiple issues ex
 ist with existing approximate inference methods ranging from estimation of
  the model order to random restarts due to the ability to converge to diff
 erent local minima. Monte-Carlo methods can be used to estimate the parame
 ters of the generating distribution and estimate the model order\, but whe
 n the distribution of each mixand has a high-dimensional parameter space\,
  they suffer from the curse of dimensionality and can take far too long to
  converge. This paper proposes a generative model for time-varying mixture
  models\, tailored for mixtures of discrete-time Markov chains. A novel\, 
 deterministic inference procedure is introduced and is shown to be suitabl
 e for applications requiring real-time estimation. The method is guarantee
 d to converge to a local maximum of the posterior likelihood at each time 
 step with a computational complexity which is low enough for real-time app
 lications. As a motivating application\, we model and predict traffic patt
 erns in a transportation network. Experiments illustrate the performance o
 f the scheme and offer insights regarding tuning of the parameters of the 
 algorithm. The experiments also investigate the predictive power of the fi
 tted model compared to less complex models and demonstrate the superiority
  of the mixture model approach for prediction of traffic routes in real da
 ta.\n
DTSTART:20161104T193000Z
DTEND:20161104T193000Z
LOCATION:room 1205\, Burnside Hall\, CA\, QC\, Montreal\, H3A 0B9\, 805 rue
  Sherbrooke Ouest
SUMMARY:Sean Lawlor\, 9IÖÆ×÷³§Ãâ·Ñ
URL:/mathstat/channels/event/sean-lawlor-mcgill-univer
 sity-263865
END:VEVENT
END:VCALENDAR