A1511
Title: Truncated inverse-Levy measure representation of the beta process
Authors: Junyi Zhang - The Education University of Hong Kong (Hong Kong) [presenting]
Angelos Dassios - London Scool of Economics (United Kingdom)
Chong Zhong - The Hong Kong Polytechnic University (China)
Qiufei Yao - Bocconi University (Italy)
Abstract: The beta process is a widely used nonparametric prior in Bayesian machine learning. While various inference schemes have been developed for the beta process and related models, the current state-of-the-art method relies heavily on the stick-breaking representation with decreasing atom weights, which is available only for a special hyperparameter. In this work, we introduce the truncated inverse-Levy measure representation (TILe-Rep) that extends the decreasing atom weights representation of the beta process to general hyperparameters. The TILe-Rep fills the gap between the two previous stick-breaking representations. Moreover, it has lower truncation error than other sequential representations of the beta process and may lead to the posterior consistency property of Bayesian factor models. We demonstrate the usage of the TILe-Rep in the celebrated binary latent feature model and the beta process factor analysis model.
