conftrace_

2012 NeurIPS NeurIPS 2012

Truncation-free Online Variational Inference for Bayesian Nonparametric Models

Abstract

We present a truncation-free online variational inference algorithm for Bayesian nonparametric models. Unlike traditional (online) variational inference algorithms that require truncations for the model or the variational distribution, our method adapts model complexity on the fly. Our experiments for Dirichlet process mixture models and hierarchical Dirichlet process topic models on two large-scale data sets show better performance than previous online variational inference algorithms.

📈 Trend Setter - Federated Learning

Authors

Chong Wang, David M. Blei

Topics

Artificial Intelligence > Bayesian & Probabilistic > Bayesian Learning Artificial Intelligence > Learning Paradigms > Federated Learning Machine Learning > Learning Types > Unsupervised Learning Machine Learning > Optimization & Theory > Bayesian Inference Deep Learning > Models > Variational Inference Machine Learning > Bayesian & Probabilistic > Bayesian Learning Machine Learning > Learning Types > Online Learning Machine Learning > Bayesian & Probabilistic > Variational Inference Machine Learning > Bayesian & Probabilistic > Nonparametric Bayesian

Keywords

dirichlet process bayesian nonparametrics online learning variational inference hierarchical dirichlet process dirichlet process mixture online inference truncation-free inference bayesian nonparametric model topic model

More from NeurIPS 2012

All NeurIPS 2012 papers →

Learning Label Trees for Probabilistic Modelling of Implicit Feedback

Factoring nonnegative matrices with linear programs

Privacy Aware Learning

Provable ICA with Unknown Gaussian Noise, with Implications for Gaussian Mixtures and Autoencoders

Learning Image Descriptors with the Boosting-Trick

A latent factor model for highly multi-relational data