Online Bayesian Passive-Aggressive Learning
This work addresses the need for more efficient and flexible online learning methods in machine learning, particularly for topic modeling, though it appears incremental as an extension of existing frameworks.
The paper tackled the limitations of deterministic online Passive-Aggressive learning by introducing online Bayesian Passive-Aggressive learning, which incorporates latent variables and nonparametric Bayesian inference for more flexible analysis, resulting in significantly improved time efficiency while maintaining comparable results to batch methods on real datasets.
Online Passive-Aggressive (PA) learning is an effective framework for performing max-margin online learning. But the deterministic formulation and estimated single large-margin model could limit its capability in discovering descriptive structures underlying complex data. This pa- per presents online Bayesian Passive-Aggressive (BayesPA) learning, which subsumes the online PA and extends naturally to incorporate latent variables and perform nonparametric Bayesian inference, thus providing great flexibility for explorative analysis. We apply BayesPA to topic modeling and derive efficient online learning algorithms for max-margin topic models. We further develop nonparametric methods to resolve the number of topics. Experimental results on real datasets show that our approaches significantly improve time efficiency while maintaining comparable results with the batch counterparts.