Making Better Use of Unlabelled Data in Bayesian Active Learning

arXiv:2404.17249v1 Announce Type: new
Abstract: Fully supervised models are predominant in Bayesian active learning. We argue that their neglect of the information present in unlabelled data harms not just predictive performance but also decisions about what data to acquire. Our proposed solution is a simple framework for semi-supervised Bayesian active learning. We find it produces better-performing models than either conventional Bayesian active learning or semi-supervised learning with randomly acquired data. It is also easier to scale up than the conventional …

abstract active learning arxiv bayesian cs.lg data decisions framework information making performance predictive semi-supervised simple solution stat.ml the information type