Deep reinforced active learning for multi-class image classification
Not on the map. Why?
Abstract (arXiv)
High accuracy medical image classification can be limited by the costs of acquiring more data as well as the time and expertise needed to label existing images. In this paper, we apply active learning to medical image classification, a method which aims to maximise model performance on a minimal subset from a larger pool of data. We present a new active learning framework, based on deep reinforcement learning, to learn an active learning query strategy to label images based on predictions from a convolutional neural network. Our framework modifies the deep-Q network formulation, allowing us to pick data based additionally on geometric arguments in the latent space of the classifier, allowing for high accuracy multi-class classification in a batch-based active learning setting, enabling the agent to label datapoints that are both diverse and about which it is most uncertain. We apply our framework to two medical imaging datasets and compare with standard query strategies as well as the most recent reinforcement learning based active learning approach for image classification.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 32 of its 67 references that are in the map. First posted on arXiv 20 Jun 2022.
Builds on
32 of 67 references.
References · 32
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
- 2015Deep Reinforcement Learning with Double Q-learningcited by 1,745
- 2017Deep Bayesian Active Learning with Image Datacited by 819
- 2019Deep Batch Active Learning by Diverse, Uncertain Gradient Lower Boundscited by 610
- 2019BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learningcited by 432
- 2019Variational Adversarial Active Learningcited by 284
- 2020GLISTER: Generalization based Data Subset Selection for Efficient and Robust Learningcited by 226
- 2021Zero-Cost Proxies for Lightweight NAScited by 199
- 2017Suggestive Annotation: A Deep Active Learning Framework for Biomedical Image Segmentationcited by 146
- 2017Deep Active Learning for Named Entity Recognitioncited by 108
- 2017Learning Active Learning from Datacited by 101
20 more
- 2019Deep Active Learning: Unified and Principled Method for Query and Trainingcited by 93
- 2019Diverse mini-batch Active Learningcited by 89
- 2020Task-Aware Variational Adversarial Active Learningcited by 84
- 2017Learning how to Active Learn: A Deep Reinforcement Learning Approachcited by 79
- 2018Efficient Active Learning for Image Classification and Segmentation using a Sample Selection and Conditional Generative Adversarial Networkcited by 74
- 2020State-Relabeling Adversarial Active Learningcited by 60
- 2020Reinforced active learning for image segmentationcited by 56
- 2018modAL: A modular active learning framework for Pythoncited by 35
- 2017Learning Algorithms for Active Learningcited by 30
- 2018Meta-Learning Transferable Active Learning Policies by Deep Reinforcement Learningcited by 26
- 2017Active One-shot Learningcited by 21
- 2018MedAL: Deep Active Learning Sampling Method for Medical Image Analysiscited by 20
- 2019Deep Active Learning with Adaptive Acquisitioncited by 17
- 2018Discovering General-Purpose Active Learning Strategiescited by 11
- 2018Active Learning for Segmentation by Optimizing Content Information for Maximal Entropycited by 10
- 2019Coresets for Data-efficient Training of Machine Learning Modelscited by 9
- 2016Geometry in Active Learning for Binary and Multi-class Image Segmentationcited by 8
- 2016Can Active Learning Experience Be Transferred?cited by 5
- 2021MedSelect: Selective Labeling for Medical Image Classification Combining Meta-Learning with Deep Reinforcement Learningcited by 4
- 2018Learning a Policy for Opportunistic Active Learningcited by 3
Built on by
Citing papers per year: 2023: 2; 2024: 1.
3 papers.