BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY: Finding Complex Structure in Biological Data - Ioannis Kontoyiann
 is\, University of Cambridge and Athens University of Economics &amp\; Bus
 iness
DTSTART:20180503T130000Z
DTEND:20180503T140000Z
UID:TALK101311@talks.cam.ac.uk
CONTACT:Alberto Padoan
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 early approaches develope
 d were ineffective in practice\, because the amount of data required for r
 eliable modeling grew exponentially with memory length. On the other hand\
 , many of the more modern methodologies that make use of more flexible and
  parsimonious models\, result in algorithms that do not scale well and are
  computationally ineffective for larger data sets.\n\nWe will discuss a cl
 ass of novel methodological tools for effective Bayesian inference for gen
 eral discrete time series\, which offer promising results on both small an
 d big data. Our starting point is the development of a rich class of Bayes
 ian hierarchical models for variable-memory Markov chains. The particular 
 prior structure we adopt makes it possible to design effective\, linear-ti
 me algorithms that can compute most of the important features of the resul
 ting posterior and predictive distributions without resorting to simulatio
 n.\n\nWe have applied the resulting tools to numerous application-specific
  tasks (including on-line prediction\, segmentation\, classification\, ano
 maly detection\, entropy estimation\, and causality testing) on data sets 
 from a very broad range of applications. Results on both simulated and rea
 l data will be presented\, with an emphasis on data sets from neuroscience
  and genetics studies.
LOCATION:Cambridge University Engineering Department\, Lecture Room 10
END:VEVENT
END:VCALENDAR