Redundancy Removal Adversarial Active Learning Based on Norm Online Uncertainty Indicator
Open Access
- 1 January 2021
- journal article
- research article
- Published by Wiley in Computational Intelligence and Neuroscience
- Vol. 2021 (1)
Abstract
Active learning aims to select the most valuable unlabelled samples for annotation. In this paper, we propose a redundancy removal adversarial active learning (RRAAL) method based on norm online uncertainty indicator, which selects samples based on their distribution, uncertainty, and redundancy. RRAAL includes a representation generator, state discriminator, and redundancy removal module (RRM). The purpose of the representation generator is to learn the feature representation of a sample, and the state discriminator predicts the state of the feature vector after concatenation. We added a sample discriminator to the representation generator to improve the representation learning ability of the generator and designed a norm online uncertainty indicator (Norm‐OUI) to provide a more accurate uncertainty score for the state discriminator. In addition, we designed an RRM based on a greedy algorithm to reduce the number of redundant samples in the labelled pool. The experimental results on four datasets show that the state discriminator, Norm‐OUI, and RRM can improve the performance of RRAAL, and RRAAL outperforms the previous state‐of‐the‐art active learning methods.Keywords
Funding Information
- Natural Science Foundation of Heilongjiang Province (LH2019C003)
- Fundamental Research Funds for Central Universities of the Central South University (2572018BH07)
This publication has 30 references indexed in Scilit:
- Multiobjective Bayesian Optimization for Active Load Modulation in a Broadband 20-W GaN Doherty Power Amplifier DesignIEEE Transactions on Microwave Theory and Techniques, 2016
- Active Image Segmentation Propagation2016
- The Cityscapes Dataset for Semantic Urban Scene Understanding2016
- Scalable Semi-Supervised Learning by Efficient Anchor Graph RegularizationIEEE Transactions on Knowledge and Data Engineering, 2016
- Observation of Gravitational Waves from a Binary Black Hole MergerPhysical Review Letters, 2016
- Single-Image Super-Resolution Using Active-Sampling Gaussian Process RegressionIEEE Transactions on Image Processing, 2015
- Querying Discriminative and Representative Samples for Batch Mode Active LearningACM Transactions on Knowledge Discovery From Data, 2015
- Semi-supervised deep learning for object tracking and classification2014
- Active learning for semantic segmentation with expected change2012
- Active Learning Methods for Remote Sensing Image ClassificationIEEE Transactions on Geoscience and Remote Sensing, 2009