Semi-Supervised Learning

By Zhiwu Lu, Peng Han, Liwei Wang, Ji-rong Wen. 2014

517 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Representation & Self-Supervised Learning

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

Citations in the corpus

Big Self-Supervised Models are Strong Semi-Supervised Learners (2020, neutral): “Aside from the representation learning paradigm, there is a large and diverse set of approaches for semi-supervised learning, we refer readers to [52–54] for surveys of classical approaches.”
FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence (2020, background): “Broader introductions to the ﬁeld are provided in [51, 52, 5].”
MixMatch: A Holistic Approach to Semi-Supervised Learning (2019, background): “More comprehensive overviews are provided in [49, 6].”
Semi-supervised Learning with Ladder Networks (2015, background): “Semi-supervised learning (Chapelle et al., 2006) studies how auxiliary unlabeled data {x(n) | N + 1 ≤ n ≤ M} can help in training a classifier.”
Seeing through the Human Reporting Bias: Visual Classifiers from Noisy Human-Centric Labels (2015, methodology): “eled training data to bootstrap parameter estimation. Our model is also speciﬁcally designed to handle the noise found in human-centric annotations. Bootstrapping [40], semi-supervised learning (SSL) [45, 57] and Positive Unlabeled (PU) learning [10,28,30,34] are other ways of learning from noisy la”
Meta-Learning for Semi-Supervised Few-Shot Classification (2018, background): “As for the literature on semi-supervised learning, while it is quite vast (Zhu, 2005; Chapelle et al., 2010), the most relevant category to our work is related to self-training (Yarowsky, 1995; Rosenberg et al., 2005).”
On causal and anticausal learning (2012, methodology): “We ﬁrst analyze the results in the benchmark chapter of a book on SSL (Tables 21.11 and 21.13 of Chapelle et al. (2006)), for the case of 100 labeled training points.”
Pitfalls of Graph Neural Network Evaluation (2018, background): “Label Propagation (LabelProp) and Normalized Laplacian Label Propagation (LabelProp NL) [Chapelle et al., 2009], on the other hand, only consider the graph structure and ignore the node attributes.”
Zero-Shot Learning—A Comprehensive Evaluation of the Good, the Bad and the Ugly (2017, background): “In zero-shot learning, transductive setting [69], [70] implies that unlabeled images from unseen classes are available during training.”
Data Distillation: Towards Omni-Supervised Learning (2017, background): “There is a great volume of work on semi-supervised learning, and comprehensive surveys can be found in [49, 4, 50].”
Unlabeled Data Improves Adversarial Robustness (2019, methodology): “We note that there are many additional approaches to semisupervised learning, including transductive SVMs, graph-based methods, and generative modeling [8, 58].”
Are Labels Required for Improving Adversarial Robustness? (2019, background): “The semisupervised learning approach [11] which, in addition to labeled data, also uses unlabeled data to learn better models is particularly relevant to our work.”
From Softmax to Sparsemax: A Sparse Model of Attention and Multi-Label Classification (2016, background): “(4)This scenario is also relevant for “learning with a probabilistic teacher” (Agrawala, 1970) and semi-supervised learning (Chapelle et al., 2006), as it can model label uncertainty.”
Co-teaching: Robust training of deep neural networks with extremely noisy labels (2018, neutral): “Nevertheless, the idea of self-paced MentorNet is similar to the self-training approach [6], and it inherited the same inferiority of accumulated error caused by the sample-selection bias.”
Adversarial Attacks on Neural Networks for Graph Data (2018, background): “While many classical approaches have been introduced in the past to tackle the node classification problem [8, 22], the last years have seen a tremendous interest in methods for deep learning on graphs [5, 7, 26].”