Semi-Supervised Learning with Variational Bayesian Inference and Maximum Uncertainty Regularization

We propose two generic methods for improving semi-supervised learning (SSL). The first integrates weight perturbation (WP) into existing “consistency regularization” (CR) based methods. implement WP by leveraging variational Bayesian inference (VBI). second method proposes a novel consistency loss called “maximum uncertainty (MUR). While most losses act on perturbations in the vicinity of each data point, MUR actively searches “virtual” points situated beyond this region that cause uncertain class predictions. This allows to impose smoothness wider area input-output manifold. Our experiments show clear improvements classification errors various CR when they are combined with VBI or both.