BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Small Big Data: Temporal structure in discrete time series - Ioann
 is Kontoyiannis (University of Cambridge\; Athens University of Economics 
 and Business)
DTSTART:20180118T114500Z
DTEND:20180118T123000Z
UID:TALK97798@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:The identification of useful temporal structure in discrete ti
 me series is an important component of algorithms used for many tasks in s
 tatistical inference and machine learning. Most of the early approaches de
 veloped were ineffective in practice\, because the amount of data required
  for reliable modeling grew exponentially with memory length. On the other
  hand\, many of the more modern methodological approaches that make use of
  more flexible and parsimonious models result in algorithms that do not sc
 ale well and are computationally ineffective for larger data sets. <br><br
 >We will discuss a class of novel methodological tools for effective Bayes
 ian inference and model selection for general discrete time series\, which
  offer promising results on both small and big data. Our starting point is
  the development of a rich class of Bayesian hierarchical models for varia
 ble-memory Markov chains. The particular prior structure we adopt makes it
  possible to design effective\, linear-time algorithms that can compute mo
 st of the important features of the resulting posterior and predictive dis
 tributions without resorting to MCMC.  <br><span><br>We have applied the r
 esulting algorithms and methodological tools to numerous application-speci
 fic tasks (including on-line prediction\, segmentation\, classification\, 
 anomaly detection\, entropy estimation\, and causality testing) on data se
 ts from different areas of application\, including data compression\, neur
 oscience\, finance\, genetics\, and animal communication. Results on both 
 simulated and real data will be presented\, and brief comparisons with oth
 er approaches (including B&uuml\;hlmann et al&rsquo\;s VLMC\, Ron et al&rs
 quo\;s PST\, and Raftery&rsquo\;s MTD) will be discussed.&nbsp\;</span>
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR