Learning with Local and Global Consistency

By Dengyong Zhou, Olivier Bousquet, T. N. Lal, Jason Aaron Edward Weston, Bernhard Schölkopf. 2003

4,672 citations recorded by the source corpus.

Subfields: 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 Representation & Self-Supervised Learning

Citations in the corpus

Semi-Supervised Classification with Graph Convolutional Networks (2016, neutral): “…semi-supervised learning, where label information is smoothed over the graph via some form of explicit graph-based regularization (Zhu et al., 2003; Zhou et al., 2004; Belkin et al., 2006; Weston et al., 2012), e.g. by using a graph Laplacian regularization term in the loss function: (1) Here, L 0…”
Revisiting Semi-Supervised Learning with Graph Embeddings (2016, background): “Graph-based semi-supervised learning deﬁnes the loss function as a weighted sum of the supervised loss over labeled instances and a graph Laplacian regularization term (Zhu et al., 2003; Zhou et al., 2004; Belkin et al., 2006; Weston et al., 2012).”
Diffusion-Convolutional Neural Networks (2015, methodology): “mproves up until two hops, then rapidly degrades. We found that the SCNN outperformed both a simple logistic regression baseline; label spreading (Lbl Spr), a baseline semi-supervised learning method Zhou et al. (2004); and count-link (Cnt Lnk), a probabilistic relational model that had been used in”
Seeing through the Human Reporting Bias: Visual Classifiers from Noisy Human-Centric Labels (2015, background): “supervised learning (SSL) approaches [60] are other ways of learning from noisy labeled data. However, they require access to clean oracle labels. SSL approaches are often computationally impractical [59,61] or make strong independence assumptions [14] that do not hold in human-centric annotations. ”
Decentralized Collaborative Learning of Personalized Models over Networks (2016, methodology): “Our ﬁrst approach, called model propagation , is inspired by the graph-based label propagation technique of Zhou et al. (2004).”
Semi-supervised learning, causality, and the conditional cluster assumption (2019, methodology): “We compare with T-SVMs [8, 16] with linear and RBF kernels using the q3svm implementation [26], and with label propagation [14, 15] using the implementation in sci-kit learn [27].”
Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learning (2018, background): “2004), spectral graph transducer (Joachims 2003), label propagation (Zhu, Ghahramani, and Lafferty 2003) and its variants (Zhou et al. 2004; Bengio, Delalleau, and Le Roux 2006), modified adsorption (Talukdar and Crammer 2009), and iterative classification algorithm (Sen et al.”
PTE: Predictive Text Embedding through Large-scale Heterogeneous Text Networks (2015, methodology): “In the setting of semisupervised learning, we also compare with classical semisupervised approaches, Naive Bayes with EM (NB+EM) [25] and label propagation (LP) [31].”
Simplifying Graph Convolutional Networks (2019, background): “Graph Laplacian regularization (Zhu et al., 2003; Zhou et al., 2004; Belkin & Niyogi, 2004; Belkin et al., 2006) introduce an additional regularization term based on graph structure which forces nodes to have similar labels to their neighbors and helps the models generalize.”
Attention-based Graph Neural Network for Semi-supervised Learning (2018, methodology): “…and un-labled data has been solved using two different approaches - Graph Laplacian based algorithms solving for locally consistent solutions [47] and Expectation Maximization based algorithms [33] where true-labels of the unlabeled data points are considered as the latent variables of a…”
The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains (2012, background): “…literature on graph-based data analysis techniques em-anates from the statistical learning community, as graph-based methods became especially popular for the semi-supervised learning problem where the objective is to classify unknown data with the help of a few labelled samples (e.g., [4]–[9]).”
A Baseline for Few-Shot Image Classification (2019, methodology): “Transductive learning: This approach is more efficient at using few labeled data than supervised learning (Joachims, 1999; Zhou et al., 2004; Vapnik, 2013).”
MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognition (2016, background): “For example, people could adopt one of the cutting-edge unsupervised/semi-supervised clustering algorithms [21] [22] [23] [24] on our training dataset, and/or develop new algorithms which can accurately locate and remove outliers in a large, real dataset.”
Propagation kernels: efficient graph kernels from propagated information (2014, background): “Other similar update schemes, such as “label spreading” (Zhou et al. 2003), could be used in a propagation kernel as well.”
Smooth Neighbors on Teacher Graphs for Semi-Supervised Learning (2017, methodology): “Graph-based SSL methods [53, 54, 51] define the similarity of data points by a graph and make predictions smooth with respect to the graph structure.”
Semi-Supervised Learning with Normalizing Flows (2019, methodology): “We compare FlowGMM to the graph based label spreading method from Zhou et al. (2004), a Π -Model (Laine & Aila, 2016) that uses dropout perturbations, as well as supervised logistic regression, k-nearest neighbors, and a neural network trained on the labeled data only.”
Associating Genes and Protein Complexes with Disease via Network Propagation (2010, methodology): “Specifically, we use the algorithm of Zhou et al. [21] which at iteration t computes where F 1 : ~ Y .”