Learning to Detect Noisy Labels Using Model-Based Features

by   Zhihao Wang, et al.
Tsinghua University

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.


Enhanced Meta Label Correction for Coping with Label Corruption

Traditional methods for learning with the presence of noisy labels have ...

Data-driven Meta-set Based Fine-Grained Visual Classification

Constructing fine-grained image datasets typically requires domain-speci...

Meta Label Correction for Learning with Weak Supervision

Leveraging weak or noisy supervision for building effective machine lear...

Maximising the Utility of Validation Sets for Imbalanced Noisy-label Meta-learning

Meta-learning is an effective method to handle imbalanced and noisy-labe...

Noisy Label Learning for Security Defects

Data-driven software engineering processes, such as vulnerability predic...

Meta-Learning via Feature-Label Memory Network

Deep learning typically requires training a very capable architecture us...

Expertise-based Weighting for Regression Models with Noisy Labels

Regression methods assume that accurate labels are available for trainin...

Please sign up or login with your details

Forgot password? Click here to reset