Self-labeled techniques for semi-supervised learning: taxonomy, software and empirical study

By Isaac Triguero, Salvador Garcia, Francisco Herrera. 2015

586 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Representation & Self-Supervised Learning, Evaluation & Benchmarks

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

In Representation & Self-Supervised Learning

In Evaluation & Benchmarks

Citations in the corpus

A Survey on Data Collection for Machine Learning: A Big Data - AI Integration Perspective (2018, methodology): “Semi-supervised learning is a broad topic, and we focus on a smaller branch of research called self-labeled techniques [96] where the goal is to generate more labels by”
Continual Learning for Named Entity Recognition (2021, background): “Self-training (Triguero, Garc´ıa, and Herrera 2015) is the semi-supervised iterative procedure in which a model is ﬁrst trained on a small labeled dataset, and then it is used to predict labels of unlabeled examples.”
Learning to Self-Train for Semi-Supervised Few-Shot Classification (2019, methodology): “Prior to this step we also perform hard selection to only use the most confident predictions [35].”
Scalable Semi-Supervised Learning by Efficient Anchor Graph Regularization (2016, methodology): “This learning mechanism is broadly used in many real-world applications such as data mining [30], [34], [35], [36], [52], [53] and multimedia content analysis [14], [15], [24], [25], [49].”
Instance Adaptive Self-Training for Unsupervised Domain Adaptation (2020, methodology): “works iteratively train the model by using both the labeled source domain data and the generated pseudo-labels in the target domain and thus achieve the alignment between the source and target domain [26]. However, these methods directly choose pseudo-labels with high prediction condence, which will”
A survey on semi-supervised learning (2019, methodology): “A comprehensive survey of wrapper methods was published recently by Triguero et al. (2015).”
Evolutionary Fuzzy Systems for Explainable Artificial Intelligence: Why, When, What for, and Where to? (2019, background): “It is based on problems in which only a subset of the instances are labeled and the algorithm can modify the output of the training data [56].”
MIST: Multiple Instance Self-Training Framework for Video Anomaly Detection (2021, background): “Self-training has been widely investigated in semi-supervised learning [1, 10, 6, 27, 22, 35].”
A cloud detection algorithm for satellite imagery based on deep learning (2019, methodology): “This could potentially be improved further by a recursive implementation, where models are trained and used to produce improved training data (Triguero et al., 2015).”
Deep Unsupervised Domain Adaptation: A Review of Recent Advances and Perspectives (2022, background): “Whereas self-training was initially presented as part of semi-supervised learning (Triguero et al., 2015), recently proposed deep self-training methods combine feature embedding with alternative learning in a unified manner, thus yielding flexible domain adaptation (Zou et al., 2019).”
Deep Learning for Land Use and Land Cover Classification Based on Hyperspectral and Multispectral Earth Observation Data: A Review (2020, methodology): “One way is to use self-labelling techniques by using a pre-trained labelling classifier [139], and another recent way is to use Generative Adversarial Networks (GAN) including generative models together with discriminative evaluation methods [62] (shown in Figure 11).”
Cross-Domain WiFi Sensing with Channel State Information: A Survey (2022, background): “The opportunity to take advantage of self-learning [72] and active learning [58] can be further exploited in the near future.”
kNN-IS: An Iterative Spark-based design of the k-Nearest Neighbors classifier for big data (2017, background): “Another direction for future work is to extend the application of the presented kNN-IS approach to a big data semi-supervised learning [46] context.”
Data collection and quality challenges in deep learning: a data-centric AI perspective (2021, background): “learning [160,188] where the idea is to use existing labels to predict the other labels.”
Confidence Regularized Self-Training (2019, background): “An overview of different self-training techniques is presented in [59].”