Results 11 to 20 of about 12,455,841 (338)
Active learning for data streams: a survey [PDF]
Machine-mediated learning, 2023 Online active learning is a paradigm in machine learning that aims to select the most informative data points to label from a data stream. The problem of minimizing the cost associated with collecting labeled observations has gained a lot of attention in
Davide Cacciarelli, M. Kulahci
semanticscholar +1 more source
Active Learning by Acquiring Contrastive Examples [PDF]
Conference on Empirical Methods in Natural Language Processing, 2021 Common acquisition functions for active learning use either uncertainty or diversity sampling, aiming to select difficult and diverse data points from the pool of unlabeled data, respectively.
Katerina Margatina +3 more
semanticscholar +1 more source
A Survey of Deep Active Learning [PDF]
ACM Computing Surveys, 2020 Active learning (AL) attempts to maximize a model’s performance gain while annotating the fewest samples possible. Deep learning (DL) is greedy for data and requires a large amount of data supply to optimize a massive number of parameters if the model is
Pengzhen Ren +6 more
semanticscholar +1 more source
Active Learning by Feature Mixing [PDF]
Computer Vision and Pattern Recognition, 2022 The promise of active learning (AL) is to reduce labelling costs by selecting the most valuable examples to annotate from a pool of unlabelled data. Identifying these examples is especially challenging with high-dimensional data (e.g. images, videos) and
Amin Parvaneh +5 more
semanticscholar +1 more source
A Comparative Survey of Deep Active Learning [PDF]
arXiv.org, 2022 While deep learning (DL) is data-hungry and usually relies on extensive labeled data to deliver good performance, Active Learning (AL) reduces labeling costs by selecting a small proportion of samples from unlabeled data for labeling and training ...
Xueying Zhan +5 more
semanticscholar +1 more source
Active Learning Through a Covering Lens [PDF]
Neural Information Processing Systems, 2022 Deep active learning aims to reduce the annotation cost for the training of deep models, which is notoriously data-hungry. Until recently, deep active learning methods were ineffectual in the low-budget regime, where only a small number of examples are ...
Ofer Yehuda +3 more
semanticscholar +1 more source
Uncertainty-driven dynamics for active learning of interatomic potentials
Nature Computational Science, 2023 Machine learning (ML) models, if trained to data sets of high-fidelity quantum simulations, produce accurate and efficient interatomic potentials. Active learning (AL) is a powerful tool to iteratively generate diverse data sets. In this approach, the ML
M. Kulichenko +7 more
semanticscholar +1 more source
Active Learning for Domain Adaptation: An Energy-based Approach [PDF]
AAAI Conference on Artificial Intelligence, 2021 Unsupervised domain adaptation has recently emerged as an effective paradigm for generalizing deep neural networks to new target domains. However, there is still enormous potential to be tapped to reach the fully supervised performance. In this paper, we
Binhui Xie +5 more
semanticscholar +1 more source
Multiple Instance Active Learning for Object Detection [PDF]
Computer Vision and Pattern Recognition, 2021 Despite the substantial progress of active learning for image recognition, there still lacks an instance-level active learning method specified for object detection.
Tianning Yuan +6 more
semanticscholar +1 more source
Defining Active Learning: A Restricted Systemic Review
Teaching and Learning Inquiry, 2023 What is active learning? While active learning has been demonstrated to have positive impacts on student learning and performance, defining the concept has been elusive.
Peter Doolittle +2 more
semanticscholar +1 more source