paper-with-me

홈 › Papers

Learning to Detect Noisy Labels Using Model-Based Features

2022-12-28 · Zhihao Wang, Zongyu Lin, Peiqi Liu, Guidong Zheng, Junjie Wen, Xianxin Chen, Yujun Chen, Zhilin Yang

Label noise is ubiquitous in various machine learning scenarios such as self-labeling with model predictions and erroneous data annotation. Many existing approaches are based on heuristics such as sample losses, which might not be flexible enough to achieve optimal solutions. Meta learning based methods address this issue by learning a data selection function, but can be hard to optimize. In light of these pros and cons, we propose Selection-Enhanced Noisy label Training (SENT) that does not rely on meta learning while having the flexibility of being data-driven. SENT transfers the noise distribution to a clean set and trains a model to distinguish noisy labels from clean ones using model-based features. Empirically, on a wide range of tasks including text classification and speech recognition, SENT improves performance over strong baselines under the settings of self-training and label corruption.

📄 PDF Abstract BibTeX arXiv:2212.13767

Code (1)

rafa-zy/sent 공식 구현 pytorch

Tasks

Meta-Learningspeech-recognitionSpeech Recognitiontext-classificationText Classification

Similar Papers 제목 키워드 기반

ADMoE: Anomaly Detection with Mixture-of-Experts from Noisy Labels

2022-08-24 · Yue Zhao, Guoqing Zheng, Subhabrata Mukherjee, Robert McCann 외

Existing works on anomaly detection (AD) rely on clean labels from human annotators that are expensive to acquire in practice. In this work, we propose a method to leverage weak/noisy labels (e.g., risk scores generated …

Anomaly DetectionMixture-of-Experts

Iterative Learning with Open-set Noisy Labels

2018-03-31 · CVPR 2018 6 · Yisen Wang, Weiyang Liu, Xingjun Ma, James Bailey 외

Large-scale datasets possessing clean label annotations are crucial for training Convolutional Neural Networks (CNNs). However, labeling large-scale data can be very costly and error-prone, and even high-quality datasets…

Incomplete Multi-View Weak-Label Learning with Noisy Features and Imbalanced Labels

2022-01-04 · Zhiwei Li, Zijian Yang, Lu Sun, Mineichi Kudo 외

A variety of modern applications exhibit multi-view multi-label learning, where each sample has multi-view features, and multiple labels are correlated via common views. Current methods usually fail to directly deal with…

Multi-Label Learning

Scalable Penalized Regression for Noise Detection in Learning with Noisy Labels

2022-03-15 · CVPR 2022 1 · Yikai Wang, Xinwei Sun, Yanwei Fu

Noisy training set usually leads to the degradation of generalization and robustness of neural networks. In this paper, we propose using a theoretically guaranteed noisy label detection framework to detect and remove noi…

Image ClassificationLearning with noisy labelsregression

A Data-Centric Framework for Detecting and Correcting Corrupted Labels

2026-06-10 · Ha-Linh Nguyen, Hong-Anh Nguyen, Minh-Duc La, Thu-Trang Nguyen 외 arxiv

The performance of machine learning and deep learning models largely depends on the quality of the training data. However, the quality of the real-world datasets is often compromised by noisy labels, which can substantia…