Abstract:Semi-supervised learning (SSL) has been extensively studied to improve the generalization ability of deep neural networks for visual recognition. To involve the unlabelled data, most existing SSL methods are based on common density-based cluster assumption: samples lying in the same high-density region are likely to belong to the same class, including the methods performing consistency regularization or generating pseudo-labels for the unlabelled images. Despite their impressive performance, we argue three limitations exist: 1) Though the density information is demonstrated to be an important clue, they all use it in an implicit way and have not exploited it in depth. 2) For feature learning, they often learn the feature embedding based on the single data sample and ignore the neighborhood information. 3) For label-propagation based pseudo-label generation, it is often done offline and difficult to be end-to-end trained with feature learning. Motivated by these limitations, this paper proposes to solve the SSL problem by building a novel density-aware graph, based on which the neighborhood information can be easily leveraged and the feature learning and label propagation can also be trained in an end-to-end way. Specifically, we first propose a new Density-aware Neighborhood Aggregation(DNA) module to learn more discriminative features by incorporating the neighborhood information in a density-aware manner. Then a novel Density-ascending Path based Label Propagation(DPLP) module is proposed to generate the pseudo-labels for unlabeled samples more efficiently according to the feature distribution characterized by density. Finally, the DNA module and DPLP module evolve and improve each other end-to-end. Extensive experiments demonstrate the effectiveness of the newly proposed density-aware graph based SSL framework and our approach can outperform current state-of-the-art methods by a large margin.

Density-based laplacian kernels for semi-supervised learning

Spectral Kernel Learning for Semi-Supervised Classification.

Maximum Margin Based Semi-supervised Spectral Kernel Learning.

Semi-supervised classification with Laplacian multiple kernel learning.

Learning Bregman Distance Functions for Semi-Supervised Clustering

Semi-supervised kernel minimum squared error based on manifold structure

Semi-supervised weighted distance metric learning for kNN classification

Semi-supervised Laplacian regularized least squares algorithm for localization in wireless sensor networks

A Semi-supervised Kernel Learning Method Based on Label Propagation

Graph-based semi-supervised learning and spectral kernel design

Simple Nonparallel Laplacian SVM for Semi-Supervised Learning on Binary Classification Problem.

Cluster kernels for semi-supervised learning

Density-Aware Graph for Deep Semi-Supervised Visual Recognition

A Coarse‐to‐fine Two‐step Method for Semisupervised Classification Using Quasi‐linear Laplacian SVM

Local Linear Embedding Algorithms Based on Probability Distributions

Laplacian pair-weight vector projection for semi-supervised learning

Semi-supervised Learning Through Adaptive Laplacian Graph Trimming

Semi-supervised low rank kernel learning algorithm via extreme learning machine

A Semi-supervised Clustering Method Through Bottleneck Distance Exploration

Unsupervised Non-Parametric Kernel Learning Algorithm

Learning the kernel matrix by resampling.