Some works [14,15] use disagreement to refine the optimization step for noisy labels. For example, co-teaching [15] iteratively trains two models. Show abstract On better detecting and leveraging noisy samples for learning with severe label noise 2023, Pattern Recognition Citation Excerpt : Co-...
Recently, there has been growing interest in developing robust models for fine-grained classification with noisy labels. One approach is to use deep neural networks (DNNs) that are specifically designed to handle noisy labels. DNNs have shown impressive performance on various machine learning tasks, ...
Clean data is critical to the success of deep learning training. In practice, noisy labels are often included in the data. When the network learns these no
然而,将active learning整合到 deep learning 的相关研究工作却非常少。Wang and Shang 是第一个 incorporate active learning with deep learning 的研究者,他们使用的 deep leanring 模型是 stacked RBM(限制玻尔兹曼机) 和 stacked autoencoders(自编码器)。Li 等人使用相似的想法解决高光谱的图像分类问题。Stark ...
Noisy labelsDeep learningRobust lossSince annotating fine-grained labels requires special expertise, label annotations often lack quality for many real-world ... X Tan,Z Dong,H Zhao - 《Visual Computer》 被引量: 0发表: 2023年 Web-Supervised Network with Softly Update-Drop Training for Fine-Grai...
Class-Balanced Loss Based on Effective Number of Samples. CVPR 2019 computer-visiondeep-learningtensorflowimagenetcvprfine-grainedinaturalistfine-grained-classificationfine-grained-visual-categorizationcvpr2019cloud-tpu UpdatedAug 29, 2021 Python Multi-label Classification with BERT; Fine Grained Sentiment Analy...
Robust fine-grained image classification with noisy labelsThe Visual Computer - Since annotating fine-grained labels requires special expertise, label annotations often lack quality for many real-world fine-grained image classifications (FGIC). Due to the......
We describe a more traditional active learning approach for obtaining larger quantities of fine-grained data in Sect. 4, which serves as a comparison to purely using noisy data. We present extensive experiments in Sect. 5, and conclude with discussion in Sect. 6. The Unreasonable Effectiveness ...
Paper tables with annotated results for Learning from Noisy Labels with Coarse-to-Fine Sample Credibility Modeling
noisy labels, ignoring the fact that the difficulties of noisy samples are different, thus a rigid and unified data selection pipeline cannot tackle this problem well. In this paper, we first propose a coarse-to-fine robust learning method called CREMA, to handle noisy data in a divide-and-...