BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:New Approaches to Biomedical Data Modelling: An Introductory Tutor
 ial - Christopher M. Bishop (Microsoft Research\, Cambridge)
DTSTART:20091207T160000Z
DTEND:20091207T170000Z
UID:TALK21143@talks.cam.ac.uk
CONTACT:Florian Markowetz
DESCRIPTION:The development of high-throughput genomic techniques\, along 
 with the emergence of integrated electronic health records\, is creating n
 ew opportunities for biomedical data analysis and modelling. It also prese
 nts substantial challenges due to the large quantities of data\, together 
 with the need to incorporate complex biological prior knowledge. \n\nOver 
 the last five years a powerful new framework for machine learning has emer
 ged\, which exploits probabilistic graphical models to allow a deep integr
 ation of rich domain knowledge with statistical learning. Efficient new in
 ference algorithms\, based on local message-passing on the graph\, allow t
 his approach to be scaled to massive data sets. This framework has already
  achieved impressive technological successes\, and is now being applied to
  biomedical problems\, such as the integration of GWAS with detailed envir
 onmental and phenotype information\, as well as rich prior knowledge\, to 
 investigate gene-environment interactions in childhood asthma. \n\nThis in
 troductory tutorial will assume little or no previous knowledge of machine
  learning\, and will be illustrated with both toy examples and real-world 
 case studies.\n\n
LOCATION:Cancer Research UK Cambridge Research Institute\, Lecture Theatre
END:VEVENT
END:VCALENDAR