Robust Semi-supervised Representation Learning for Graph-Structured Data

Guo, Lan-Zhe; Han, Tao; Li, Yufeng

doi:10.1007/978-3-030-16142-2_11

Cited by 5 publications

(4 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As illustrated in [19], semi-supervised learning uses labelled as well as unlabeled data to perform certain learning tasks such as classification. Some methods [20], [21] which can be classes as self-training use base classifiers to obtain predictions for unlabeled data and these pseudo-labelled data are therefore used for supervised training. In [20], unlabeled images are labeled as the most confident of their predictions.…”

Section: Semi-supervised Learningmentioning

confidence: 99%

“…The unlabeled data are weighted to fit the training process since pseudo labels are not reliable enough in the beginning. [21] balances the influence of unlabeled data by giving pseudo label when the maximum prediction confidence is greater than a threshold. Co-training [22] and Tri-training [23] are extensions of self-training which learn from multiple supervised classifiers.…”

Section: Semi-supervised Learningmentioning

confidence: 99%

See 1 more Smart Citation

Boosting Facial Expression Recognition by A Semi-Supervised Progressive Teacher

Jiang,

Deng

2022

Preprint

View full text Add to dashboard Cite

In this paper, we aim to improve the performance of in-the-wild Facial Expression Recognition (FER) by exploiting semi-supervised learning. Large-scale labeled data and deep learning methods have greatly improved the performance of image recognition. However, the performance of FER is still not ideal due to the lack of training data and incorrect annotations (e.g., label noises). Among existing in-the-wild FER datasets, reliable ones contain insufficient data to train robust deep models while large-scale ones are annotated in lower quality. To address this problem, we propose a semi-supervised learning algorithm named Progressive Teacher (PT) to utilize reliable FER datasets as well as large-scale unlabeled expression images for effective training. On the one hand, PT introduces semi-supervised learning method to relieve the shortage of data in FER. On the other hand, it selects useful labeled training samples automatically and progressively to alleviate label noise. PT uses selected clean labeled data for computing the supervised classification loss and unlabeled data for unsupervised consistency loss. Experiments on widely-used databases RAF-DB and FERPlus validate the effectiveness of our method, which achieves state-of-the-art performance with accuracy of 89.57% on RAF-DB. Additionally, when the synthetic noise rate reaches even 30%, the performance of our PT algorithm only degrades by 4.37%.

show abstract

Section: Semi-supervised Learningmentioning

confidence: 99%

Section: Semi-supervised Learningmentioning

confidence: 99%

Boosting Facial Expression Recognition by A Semi-Supervised Progressive Teacher

Jiang,

Deng

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Consequently, deep SSL that introduces the superiority of deep models to the classic SSL frameworks has attracted much attention and a considerable number of deep SSL methods have been proposed. Deep SSL methods can be classified into five categories, i.e., consistency regularization methods [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], pseudo-labeling methods [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], holistic methods of consistency regularization and pseudo-labeling [45], [46], [47], [48], [49], deep generative SSL methods [50], [51], [52], [53], [54], [55] and deep graph-based SSL methods [56], [57], [58], [59], [60]. Deep SSL methods have been successful applied into various tasks such as image classification [49], object detection [61], semantic segmentation [62], text classification [63], question answering [64], etc.…”

Section: Introductionmentioning

confidence: 99%

Robust Deep Semi-Supervised Learning: A Brief Introduction

Guo¹,

Zhou²,

Li³

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Semi-supervised learning (SSL) is the branch of machine learning that aims to improve learning performance by leveraging unlabeled data when labels are insufficient. Recently, SSL with deep models has proven to be successful on standard benchmark tasks. However, they are still vulnerable to various robustness threats in real-world applications as these benchmarks provide perfect unlabeled data, while in realistic scenarios, unlabeled data could be corrupted. Many researchers have pointed out that after exploiting corrupted unlabeled data, SSL suffers severe performance degradation problems. Thus, there is an urgent need to develop SSL algorithms that could work robustly with corrupted unlabeled data. To fully understand robust SSL, we conduct a survey study. We first clarify a formal definition of robust SSL from the perspective of machine learning. Then, we classify the robustness threats into three categories: i) distribution corruption, i.e., unlabeled data distribution is mismatched with labeled data; ii) feature corruption, i.e., the features of unlabeled examples are adversarially attacked; and iii) label corruption, i.e., the label distribution of unlabeled data is imbalanced. Under this unified taxonomy, we provide a thorough review and discussion of recent works that focus on these issues. Finally, we propose possible promising directions within robust SSL to provide insights for future research.

show abstract

“…In the second stage, it usually adopts supervised fine-tuning [25] or traditional semi-supervised classifiers [14,17,29,30,16] to further enhance the discriminative capability of the learned feature and learn the final classifier. Since network pre-training is task-agnostic, the representations generated by the network pre-training are likely to be suboptimal for the ultimate classification tasks [31,19,32,33]. Nevertheless, the pre-training-based methods are less sensitive to the confirmation bias thanks to its decoupling learning scheme, which pursues the outstanding performance of each stage separately without considering the quality of pseudo labels.…”

Section: Introductionmentioning

confidence: 99%

Robust semi-supervised classification based on data augmented online ELMs with deep features

Zeng

et al. 2021

Knowledge-Based Systems

View full text Add to dashboard Cite

Robust Semi-supervised Representation Learning for Graph-Structured Data

Cited by 5 publications

References 7 publications

Boosting Facial Expression Recognition by A Semi-Supervised Progressive Teacher

Boosting Facial Expression Recognition by A Semi-Supervised Progressive Teacher

Robust Deep Semi-Supervised Learning: A Brief Introduction

Robust semi-supervised classification based on data augmented online ELMs with deep features

Contact Info

Product

Resources

About