Bayesian Pseudo Labels: Expectation Maximization for Robust and Efficient Semi-Supervised Segmentation
This paper concerns pseudo labelling in segmentation. Our contribution is fourfold. Firstly, we present a new formulation of pseudo-labelling as an Expectation-Maximization (EM) algorithm for clear statistical interpretation. Secondly, we propose a semi-supervised medical image segmentation method purely based on the original pseudo labelling, namely SegPL. We demonstrate SegPL is a competitive approach against state-of-the-art consistency regularisation based methods on semi-supervised segmentation on a 2D multi-class MRI brain tumour segmentation task and a 3D binary CT lung vessel segmentation task. The simplicity of SegPL allows less computational cost comparing to prior methods. Thirdly, we demonstrate that the effectiveness of SegPL may originate from its robustness against out-of-distribution noises and adversarial attacks. Lastly, under the EM framework, we introduce a probabilistic generalisation of SegPL via variational inference, which learns a dynamic threshold for pseudo labelling during the training. We show that SegPL with variational inference can perform uncertainty estimation on par with the gold-standard method Deep Ensemble.
Code (1)
Tasks
Image SegmentationMedical Image SegmentationPseudo LabelSegmentationSemantic SegmentationSemi-supervised Medical Image SegmentationVariational InferenceMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Expectation Maximization Pseudo Labels
In this paper, we study pseudo-labelling. Pseudo-labelling employs raw inferences on unlabelled data as pseudo-labels for self-training. We elucidate the empirical successes of pseudo-labelling by establishing a link bet…
SegmentationSemi-supervised Contrastive Outlier removal for Pseudo Expectation Maximization (SCOPE)
Semi-supervised learning is the problem of training an accurate predictive model by combining a small labeled dataset with a presumably much larger unlabeled dataset. Many methods for semi-supervised deep learning have b…
Contrastive LearningGraph-Based Uncertainty-Aware Self-Training with Stochastic Node Labeling
Self-training has become a popular semi-supervised learning technique for leveraging unlabeled data. However, the over-confidence of pseudo-labels remains a key challenge. In this paper, we propose a novel \emph{graph-ba…
Node ClassificationPseudo LabelUncertainty-Aware Graph Self-Training with Expectation-Maximization Regularization
In this paper, we propose a novel \emph{uncertainty-aware graph self-training} approach for semi-supervised node classification. Our method introduces an Expectation-Maximization (EM) regularization scheme to incorporate…
Node ClassificationPseudo LabelA Pseudo Label based Dataless Naive Bayes Algorithm for Text Classification with Seed Words
Traditional supervised text classifiers require a large number of manually labeled documents, which are often expensive to obtain. Recently, dataless text classification has attracted more attention, since it only requir…
General ClassificationPseudo Labeltext-classificationText Classification