PASS: Peer-Agreement based Sample Selection for training with Noisy Labels

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Garg, Arpit, Nguyen, Cuong, Felix, Rafael, Do, Thanh-Toan, Carneiro, Gustavo
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917190973784064
author Garg, Arpit
Nguyen, Cuong
Felix, Rafael
Do, Thanh-Toan
Carneiro, Gustavo
author_facet Garg, Arpit
Nguyen, Cuong
Felix, Rafael
Do, Thanh-Toan
Carneiro, Gustavo
contents The prevalence of noisy-label samples poses a significant challenge in deep learning, inducing overfitting effects. This has, therefore, motivated the emergence of learning with noisy-label (LNL) techniques that focus on separating noisy- and clean-label samples to apply different learning strategies to each group of samples. Current methodologies often rely on the small-loss hypothesis or feature-based selection to separate noisy- and clean-label samples, yet our empirical observations reveal their limitations, especially for labels with instance dependent noise (IDN). An important characteristic of IDN is the difficulty to distinguish the clean-label samples that lie near the decision boundary (i.e., the hard samples) from the noisy-label samples. We, therefore, propose a new noisy-label detection method, termed Peer-Agreement based Sample Selection (PASS), to address this problem. Utilising a trio of classifiers, PASS employs consensus-driven peer-based agreement of two models to select the samples to train the remaining model. PASS is easily integrated into existing LNL models, enabling the improvement of the detection accuracy of noisy- and clean-label samples, which increases the classification accuracy across various LNL benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2303_10802
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle PASS: Peer-Agreement based Sample Selection for training with Noisy Labels
Garg, Arpit
Nguyen, Cuong
Felix, Rafael
Do, Thanh-Toan
Carneiro, Gustavo
Computer Vision and Pattern Recognition
The prevalence of noisy-label samples poses a significant challenge in deep learning, inducing overfitting effects. This has, therefore, motivated the emergence of learning with noisy-label (LNL) techniques that focus on separating noisy- and clean-label samples to apply different learning strategies to each group of samples. Current methodologies often rely on the small-loss hypothesis or feature-based selection to separate noisy- and clean-label samples, yet our empirical observations reveal their limitations, especially for labels with instance dependent noise (IDN). An important characteristic of IDN is the difficulty to distinguish the clean-label samples that lie near the decision boundary (i.e., the hard samples) from the noisy-label samples. We, therefore, propose a new noisy-label detection method, termed Peer-Agreement based Sample Selection (PASS), to address this problem. Utilising a trio of classifiers, PASS employs consensus-driven peer-based agreement of two models to select the samples to train the remaining model. PASS is easily integrated into existing LNL models, enabling the improvement of the detection accuracy of noisy- and clean-label samples, which increases the classification accuracy across various LNL benchmarks.
title PASS: Peer-Agreement based Sample Selection for training with Noisy Labels
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2303.10802