BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Dynamic causal networks with multi-scale temporal structure - Eric
  Kolaczyk (Boston University)
DTSTART:20161212T111500Z
DTEND:20161212T120000Z
UID:TALK69451@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:<span>Co-authors: Xinyu Kang (Boston University)\, Apratim  Ga
 nguly (Boston University) <br></span> <span><br>I will discuss a novel met
 hod to model multivariate time series using dynamic  causal networks. This
  method combines traditional multi-scale modeling and  network based neigh
 borhood selection\, aiming at capturing the temporally local  structure of
  the data while maintaining the sparsity of the potential  interactions. O
 ur multi-scale framework is based on recursive dyadic  partitioning\, whic
 h recursively partitions the temporal axis into finer  intervals and allow
 s us to detect local network structural changes at varying  temporal resol
 utions. The dynamic neighborhood selection is achieved through  penalized 
 likelihood estimation\, where the penalty seeks to limit the number of  ne
 ighbors used to model the data. Theoretical and numerical results describi
 ng  the performance of our method will be presented\, as well as an applic
 ation in  computational neuroscience.&nbsp\;</span>
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR