Belief Propagation Algorithms with Applications to Cancer Genomics

Belief propagation is an algorithmic method that provides solutions to many complex machine-learning problems.

Belief propagation algorithms have numerous applications in coding theory, vision and many other areas of machine learning. In this talk, Christian Borgs will give an introduction to belief propagation, discuss how the accuracy of belief propagation has been rigorously established, and present recent applications to systems biology. Examples include simple applications to yeast networks, complex applications in the discovery of pathways in cancer genomics, and modifications to distinguish patient-specific pathways from more general disease pathways.

Christian Borgs is cofounder of Microsoft Research in Cambridge, Massachusetts, where he serves as principal researcher and deputy managing director. He is known for the use of statistical physics methods in computer science, and focuses on the science of networks, including mathematical foundations, graph algorithms, applications in economics, and systems biology. He is a Fellow of the American Mathematical Society and the American Association for the Advancement of Science.