Active Machine Learning for Embryogenesis

old_uid5319
titleActive Machine Learning for Embryogenesis
start_date2008/10/02
schedule17h
onlineno
summaryThe intrinsic complexity of biological systems creates huge amounts of unlabeled experimental data. Processing and analyzing this data can be assisted by active machine learning (AML), in which automatic classification is combined with human expertise. Experts define high-level symbolic categories and tentative class boundaries based on sparse data. We present a global strategy for designing AML methods suited for the observation and analysis of complex systems, such as embryonic development. We have developed a procedure that exploits available knowledge, whether gathered manually or automatically, and is able to adapt when new data is provided. We show that it can be a powerful method for the investigation of the morphogenetic features of embryogenesis. It can contribute to the reconstruction of in vivo cell morphodynamics, one of the main challenges of the post-genomic era.
oncancelJour et horaire inhabituels
responsiblesBourgine