A0973
Title: Bayesian structure selection approaches for multiple binary responses via multi-task learning
Authors: Chi-Hsiang Chu - Tunghai University (Taiwan) [presenting]
Abstract: Bayesian structure selection problems for the categorical response are addressed. It is focused on solving the selection problem for multiple binary responses, and the probit model for each response is used. Here, the group structure is considered with sparsity property on the rows of the coefficient matrix where each row corresponds to one variable. Then the relevant variables for the responses are identified, and the selection problems can be treated as the multi-task learning problem. The effectiveness of the proposed method will be demonstrated through simulation studies.
