Influential Rank: A New Perspective of Post-training for Robust Model against Noisy Labels

Deep neural network can easily overfit to even noisy labels due to its high capacity, which degrades the generalization performance of a model. To overcome this issue, we propose a new approach for learning from noisy labels (LNL) via post-training, which can significantly improve the generalization...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2023-04
Main Authors: Park, Seulki, Song, Hwanjun, Um, Daeho, Jo, Dae Ung, Sangdoo Yun, Jin Young Choi
Format: Article
Language:English
Subjects:
Algorithms
Artificial neural networks
Criteria
Data points
Datasets
Iterative algorithms
Iterative methods
Machine learning
Regularization
Regularization methods
Statistical methods
Training
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Deep neural network can easily overfit to even noisy labels due to its high capacity, which degrades the generalization performance of a model. To overcome this issue, we propose a new approach for learning from noisy labels (LNL) via post-training, which can significantly improve the generalization performance of any pre-trained model on noisy label data. To this end, we rather exploit the overfitting property of a trained model to identify mislabeled samples. Specifically, our post-training approach gradually removes samples with high influence on the decision boundary and refines the decision boundary to improve generalization performance. Our post-training approach creates great synergies when combined with the existing LNL methods. Experimental results on various real-world and synthetic benchmark datasets demonstrate the validity of our approach in diverse realistic scenarios.
ISSN:2331-8422