Adaptive Negative Evidential Deep Learning for Open-set Semi-supervised Learning
Preprint
Abstract
Semi-supervised learning (SSL) methods assume that labeled data, unlabeled data and test data are from the same distribution. Open-set semi-supervised learning (Open-set SSL) considers a more practical scenario, where unlabeled data and test data contain new categories (outliers) not observed in labeled data (inliers). Most previous works focused on outlier detection via binary classifiers, which suffer from insufficient scalability and inability to distinguish different types of uncertainty. In this paper, we propose a novel framework, Adaptive Negative Evidential Deep Learning (ANEDL) to tackle these limitations. Concretely, we first introduce evidential deep learning (EDL) as an outlier detector to quantify different types of uncertainty, and design different uncertainty metrics for self-training and inference. Furthermore, we propose a novel adaptive negative optimization strategy, making EDL more tailored to the unlabeled dataset containing both inliers and outliers. As demonstrated empirically, our proposed method outperforms existing state-of-the-art methods across four datasets.Keywords
All Related Versions
This publication has 5 references indexed in Scilit:
- Trash to Treasure: Harvesting OOD Data with Cross-Modal Matching for Open-Set Semi-Supervised Learning2021
- Evidential Deep Learning for Guided Molecular Property Prediction and DiscoveryACS Central Science, 2021
- Multiple instance learning: A survey of problem characteristics and applicationsPattern Recognition, 2018
- Wide Residual Networks2016
- Combination of Evidence in Dempster-Shafer Theory2002