paper-with-me

홈 › Papers

HyperMatch: Noise-Tolerant Semi-Supervised Learning via Relaxed Contrastive Constraint

2023-01-01 · CVPR 2023 1 · Beitong Zhou, Jing Lu, Kerui Liu, Yunlu Xu, Zhanzhan Cheng, Yi Niu

Recent developments of the application of Contrastive Learning in Semi-Supervised Learning (SSL) have demonstrated significant advancements, as a result of its exceptional ability to learn class-aware cluster representations and the full exploitation of massive unlabeled data. However, mismatched instance pairs caused by inaccurate pseudo labels would assign an unlabeled instance to the incorrect class in feature space, hence exacerbating SSL's renowned confirmation bias. To address this issue, we introduced a novel SSL approach, HyperMatch, which is a plug-in to several SSL designs enabling noise-tolerant utilization of unlabeled data. In particular, confidence predictions are combined with semantic similarities to generate a more objective class distribution, followed by a Gaussian Mixture Model to divide pseudo labels into a 'confident' and a 'less confident' subset. Then, we introduce Relaxed Contrastive Loss by assigning the 'less-confident' samples to a hyper-class, i.e. the union of top-K nearest classes, which effectively regularizes the interference of incorrect pseudo labels and even increases the probability of pulling a 'less confident' sample close to its true class. Experiments and in-depth studies demonstrate that HyperMatch delivers remarkable state-of-the-art performance, outperforming FixMatch on CIFAR100 with 400 and 2500 labeled samples by 11.86% and 4.88%, respectively.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Contrastive Learning

Methods 이 논문이 사용한 방법론

FixMatch FixMatch is an algorithm that first generates pseudo-labels using the model's predictions on weakly-augmented unlabeled images. For a given image, the pseudo-label is only…
Contrastive Learning 설명 없음

Similar Papers 제목 키워드 기반

Simplifying Adversarially Robust PAC Learning with Tolerance

2025-02-11 · Hassan Ashtiani, Vinayak Pathak, Ruth Urner

Adversarially robust PAC learning has proved to be challenging, with the currently best known learners [Montasser et al., 2021a] relying on improper methods based on intricate compression schemes, resulting in sample com…

PAC learning

NOTE-RCNN: NOise Tolerant Ensemble RCNN for Semi-Supervised Object Detection

2018-12-01 · ICCV 2019 10 · JIyang Gao, Jiang Wang, Shengyang Dai, Li-Jia Li 외

The labeling cost of large number of bounding boxes is one of the main challenges for training modern object detectors. To reduce the dependence on expensive bounding box annotations, we propose a new semi-supervised obj…

Objectobject-detectionObject DetectionSemi-Supervised Object Detection+1

Countering Noisy Labels By Learning From Auxiliary Clean Labels

2019-05-23 · Tsung Wei Tsai, Chongxuan Li, Jun Zhu

We consider the learning from noisy labels (NL) problem which emerges in many real-world applications. In addition to the widely-studied synthetic noise in the NL literature, we also consider the pseudo labels in semi-su…

Pseudo Contrastive Learning for Graph-based Semi-supervised Learning

2023-02-19 · Weigang Lu, Ziyu Guan, Wei Zhao, Yaming Yang 외

Pseudo Labeling is a technique used to improve the performance of semi-supervised Graph Neural Networks (GNNs) by generating additional pseudo-labels based on confident predictions. However, the quality of generated pseu…

Contrastive LearningData Augmentation

Practical Validity Conditions for Byzantine-Tolerant Federated Learning

2026-05-15 · Mélanie Cambus, Darya Melnyk, Tijana Milentijević, Stefan Schmid arxiv

Robust aggregation is the core operation in Byzantine-tolerant federated learning. To ensure the quality of aggregation independently of data distribution or attacks, validity conditions are needed. They provide geometri…

Federated Learning