Active Learning by Querying Informative and Representative Examples

By Sheng-Jun Huang, Rong Jin, Zhi-Hua Zhou. 2010

731 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Machine Unlearning (2019, background): “However, not all of this data is useful [49].”
Learning Active Learning from Data (2017, background): “[12] (a strategy that uses the topology of the feature space in the query selection).”
Deep Class-Incremental Learning: A Survey (2023, positive): “The exemplar sampling process is also similar to the active learning protocol [120], and some works propose corresponding sampling measures to select the informative ones.”
A brief introduction to weakly supervised learning (2018, background): “, informativeness and representativeness [43].”
Deep learning approach for active classification of electrocardiogram signals (2016, methodology): “In order to select the most relevant beats, this work explores two different selection criteria based on entropy [21,28,48] and BT [11,39,49], respectively.”
An Overview of Deep Semi-Supervised Learning (2020, methodology): “The two widely used selection criteria are informativeness and representativeness [72, 188].”
Submodularity in Data Subset Selection and Active Learning (2015, background): “In order to obtain the benefits of both classes of algorithms, several papers have combined both notions (informativeness and representativeness) in a single objective (Xu et al., 2003; Huang et al., 2010).”
Deep Batch Active Learning by Diverse, Uncertain Gradient Lower Bounds (2019, methodology): “[28] Sheng-Jun Huang, Rong Jin, and Zhi-Hua Zhou.”
Active and Semi-Supervised Learning in ASR: Benefits on the Acoustic and Language Models (2016, methodology): “In [5], a min-max framework for selecting utterances considering both informativeness and representativeness criteria was proposed.”
Deep forest (2017, background): “Indeed, the most important value of this paper lies in the fact that it may open the door for non-NN-style deep learning, or deep models based on non-differentiable modules that do not rely on gradients.”
How to Train Your MAML to Excel in Few-Shot Classification (2021, background): “Training a model under data budgets is important in machine learning, computer vision, and many other application fields, since the costs of collecting data [55] and labeling them [43] are by no means negligible.”
Few-shot learning with adaptively initialized task optimizer: a practical meta-learning approach (2019, background): “Considering both the data collection (Li and Zhou 2015) and labeling cost (Huang et al. 2014), it is necessary to enable a machine learning algorithm to have such a valuable ability—building classiﬁers with limited training examples, i.e., the one-shot and the few-shot Learning.”
Cost-Effective Incremental Deep Model: Matching Model Capacity With the Least Sampling (2023, positive): “Traditional active learning is designed for reducing the labeling cost [16], [17], which tries to train an effective model with less queries to reduce the overall cost.”