# Uncertainty Based Detection and Relabeling of Noisy Image Labels

arXiv: https://arxiv.org/abs/1906.11876 · HTML: https://mlsanity.com/paper/1906.11876/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 May 2019.

## Built on by (most cited first, 10 of 10)

- [Addressing out-of-distribution label noise in webly-labelled data](https://mlsanity.com/paper/2110.13699/) (2021)
- [Label Cleaning Multiple Instance Learning: Refining Coarse Annotations on Single Whole-Slide Images](https://mlsanity.com/paper/2109.10778/) (2021)
- [On the Robustness of Monte Carlo Dropout Trained with Noisy Labels](https://mlsanity.com/paper/2103.12002/) (2021)
- [Data Cleaning and Machine Learning: A Systematic Literature Review](https://mlsanity.com/paper/2310.01765/) (2023)
- [Uncertainty-based method for improving poorly labeled segmentation datasets](https://mlsanity.com/paper/2102.08021/) (2021)
- [W2WNet: a two-module probabilistic Convolutional Neural Network with embedded data cleansing functionality](https://mlsanity.com/paper/2103.13107/) (2021)
- [VOICE: Variance of Induced Contrastive Explanations to quantify Uncertainty in Neural Network Interpretability](https://mlsanity.com/paper/2406.00573/) (2024)
- [Bridging Generative and Discriminative Noisy-Label Learning via Direction-Agnostic EM Formulation](https://mlsanity.com/paper/2308.01184/) (2023)
- [Selfish Evolution: Making Discoveries in Extreme Label Noise with the Help of Overfitting Dynamics](https://mlsanity.com/paper/2412.00077/) (2024)
- [Anti-Backdoor Coreset Selection via Cumulative Entropy](https://mlsanity.com/paper/2607.25502/) (2026)

## Abstract (arXiv)

Deep neural networks (DNNs) are powerful tools in computer vision tasks. However, in many realistic scenarios label noise is prevalent in the training images, and overfitting to these noisy labels can significantly harm the generalization performance of DNNs. We propose a novel technique to identify data with noisy labels based on the different distributions of the predictive uncertainties from a DNN over the clean and noisy data. Additionally, the behavior of the uncertainty over the course of training helps to identify the network weights which best can be used to relabel the noisy labels. Data with noisy labels can therefore be cleaned in an iterative process. Our proposed method can be easily implemented, and shows promising performance on the task of noisy label detection on CIFAR-10 and CIFAR-100.