Learn to Propagate Reliably on Noisy Affinity Graphs
- URL: http://arxiv.org/abs/2007.08802v1
- Date: Fri, 17 Jul 2020 07:55:59 GMT
- Title: Learn to Propagate Reliably on Noisy Affinity Graphs
- Authors: Lei Yang, Qingqiu Huang, Huaiyi Huang, Linning Xu, Dahua Lin
- Abstract summary: Recent works have shown that exploiting unlabeled data through label propagation can substantially reduce the labeling cost.
How to propagate labels reliably, especially on a dataset with unknown outliers, remains an open question.
We propose a new framework that allows labels to be propagated reliably on large-scale real-world data.
- Score: 69.97364913330989
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Recent works have shown that exploiting unlabeled data through label
propagation can substantially reduce the labeling cost, which has been a
critical issue in developing visual recognition models. Yet, how to propagate
labels reliably, especially on a dataset with unknown outliers, remains an open
question. Conventional methods such as linear diffusion lack the capability of
handling complex graph structures and may perform poorly when the seeds are
sparse. Latest methods based on graph neural networks would face difficulties
on performance drop as they scale out to noisy graphs. To overcome these
difficulties, we propose a new framework that allows labels to be propagated
reliably on large-scale real-world data. This framework incorporates (1) a
local graph neural network to predict accurately on varying local structures
while maintaining high scalability, and (2) a confidence-based path scheduler
that identifies outliers and moves forward the propagation frontier in a
prudent way. Experiments on both ImageNet and Ms-Celeb-1M show that our
confidence guided framework can significantly improve the overall accuracies of
the propagated labels, especially when the graph is very noisy.
Related papers
- ERASE: Error-Resilient Representation Learning on Graphs for Label Noise
Tolerance [53.73316938815873]
We propose a method called ERASE (Error-Resilient representation learning on graphs for lAbel noiSe tolerancE) to learn representations with error tolerance.
ERASE combines prototype pseudo-labels with propagated denoised labels and updates representations with error resilience.
Our method can outperform multiple baselines with clear margins in broad noise levels and enjoy great scalability.
arXiv Detail & Related papers (2023-12-13T17:59:07Z) - CONVERT:Contrastive Graph Clustering with Reliable Augmentation [110.46658439733106]
We propose a novel CONtrastiVe Graph ClustEring network with Reliable AugmenTation (CONVERT)
In our method, the data augmentations are processed by the proposed reversible perturb-recover network.
To further guarantee the reliability of semantics, a novel semantic loss is presented to constrain the network.
arXiv Detail & Related papers (2023-08-17T13:07:09Z) - Learning on Graphs under Label Noise [5.909452203428086]
We develop a novel approach dubbed Consistent Graph Neural Network (CGNN) to solve the problem of learning on graphs with label noise.
Specifically, we employ graph contrastive learning as a regularization term, which promotes two views of augmented nodes to have consistent representations.
To detect noisy labels on the graph, we present a sample selection technique based on the homophily assumption.
arXiv Detail & Related papers (2023-06-14T01:38:01Z) - All Points Matter: Entropy-Regularized Distribution Alignment for
Weakly-supervised 3D Segmentation [67.30502812804271]
Pseudo-labels are widely employed in weakly supervised 3D segmentation tasks where only sparse ground-truth labels are available for learning.
We propose a novel learning strategy to regularize the generated pseudo-labels and effectively narrow the gaps between pseudo-labels and model predictions.
arXiv Detail & Related papers (2023-05-25T08:19:31Z) - Feature propagation as self-supervision signals on graphs [0.0]
Regularized Graph Infomax (RGI) is a simple yet effective framework for node level self-supervised learning.
We show that RGI can achieve state-of-the-art performance regardless of its simplicity.
arXiv Detail & Related papers (2023-03-15T14:20:06Z) - Informative Pseudo-Labeling for Graph Neural Networks with Few Labels [12.83841767562179]
Graph Neural Networks (GNNs) have achieved state-of-the-art results for semi-supervised node classification on graphs.
The challenge of how to effectively learn GNNs with very few labels is still under-explored.
We propose a novel informative pseudo-labeling framework, called InfoGNN, to facilitate learning of GNNs with extremely few labels.
arXiv Detail & Related papers (2022-01-20T01:49:30Z) - Deep Fraud Detection on Non-attributed Graph [61.636677596161235]
Graph Neural Networks (GNNs) have shown solid performance on fraud detection.
labeled data is scarce in large-scale industrial problems, especially for fraud detection.
We propose a novel graph pre-training strategy to leverage more unlabeled data.
arXiv Detail & Related papers (2021-10-04T03:42:09Z) - Unified Robust Training for Graph NeuralNetworks against Label Noise [12.014301020294154]
We propose a new framework, UnionNET, for learning with noisy labels on graphs under a semi-supervised setting.
Our approach provides a unified solution for robustly training GNNs and performing label correction simultaneously.
arXiv Detail & Related papers (2021-03-05T01:17:04Z) - Temporal Calibrated Regularization for Robust Noisy Label Learning [60.90967240168525]
Deep neural networks (DNNs) exhibit great success on many tasks with the help of large-scale well annotated datasets.
However, labeling large-scale data can be very costly and error-prone so that it is difficult to guarantee the annotation quality.
We propose a Temporal Calibrated Regularization (TCR) in which we utilize the original labels and the predictions in the previous epoch together.
arXiv Detail & Related papers (2020-07-01T04:48:49Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.