BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Statistical Models and Their Applications in Biomarker Discovery -
  Antonella Iuliano\, Department of Health Science\, University of Basilica
 ta\, Italy
DTSTART:20241127T173000Z
DTEND:20241127T181500Z
UID:TALK225130@talks.cam.ac.uk
CONTACT:Pietro Lio
DESCRIPTION:n these lectures I propose and explore novel statistical model
 ing approaches\, focusing on the\nAccelerated Failure Time (AFT) model com
 bined with a penalized likelihood method for variable\nselection and estim
 ation.\nLecture 1. By using a log-linear representation\, the inference pr
 oblem is transformed into a\nstructured sparse regression problem. Specifi
 cally\, we incorporate a double penalty in the model\nthat promotes both s
 parsity and a grouping effect. We establish the theoretical consistency of
  the\nestimator and present an efficient iterative computational algorithm
  based on the proximal gradient\ndescent method.\nLecture 2. An extension 
 of this approach is the cooperative version\, which introduces an\n“agre
 ement” penalty to encourage alignment of predictions from different data
  views. This can be\nparticularly powerful when the data views share an un
 derlying relationship in their signals which\ncan be leveraged to enhance 
 the overall signal quality. We establish the theoretical consistency\nof t
 he estimator and present an efficient iterative computational algorithm ba
 sed on the proximal\ngradient descent method.\nLecture 3. We evaluate the 
 model’s performance using both synthetic data and real-world\ndata from 
 cancer survival analysis to identify potential new biomarkers.\n1
LOCATION:Lecture Theatre 2
END:VEVENT
END:VCALENDAR