Learning Parameterized Quadratic Pseudo-Boolean Functions
Stephen Gould
ARTIFICIAL INTELLIGENCE SEMINARDATE: 2012-11-30
TIME: 15:00:00 - 16:00:00
LOCATION: RSISE Seminar Room, ground floor, building 115, cnr. North and Daley Roads, ANU
CONTACT: JavaScript must be enabled to display this email address.
ABSTRACT:
Conditional Markov random fields (CRFs) are an important class of graphical model that are pervasive in computer vision and other machine learning applications. In the case of binary variables and pairwise interactions the CRF is directly related to so-called pseudo-Boolean functions. In this talk I will discuss this connection and present work on learning the parameters of a pseudo-Boolean function and hence binary pairwise CRF from labelled training data.
