Establishment of Neural Networks Robust to Label Noise

TL;DR

This paper investigates methods to improve the robustness of neural networks against label noise, focusing on transition matrix estimation and testing on CNN models with FashionMNIST datasets.

Contribution

It introduces a transition matrix estimator for label noise correction and evaluates its effectiveness on CNN classifiers, highlighting areas for further refinement.

Findings

Transition matrix estimator shows promise against actual transition matrix.

Both LeNet and AlexNet models exhibit robustness on FashionMNIST datasets.

Limited due to resource constraints, the impact of noise correction on robustness remains inconclusive.

Abstract

Label noise is a significant obstacle in deep learning model training. It can have a considerable impact on the performance of image classification models, particularly deep neural networks, which are especially susceptible because they have a strong propensity to memorise noisy labels. In this paper, we have examined the fundamental concept underlying related label noise approaches. A transition matrix estimator has been created, and its effectiveness against the actual transition matrix has been demonstrated. In addition, we examined the label noise robustness of two convolutional neural network classifiers with LeNet and AlexNet designs. The two FashionMINIST datasets have revealed the robustness of both models. We are not efficiently able to demonstrate the influence of the transition matrix noise correction on robustness enhancements due to our inability to correctly tune the complex convolutional neural network model due to time and computing resource constraints. There is a need for additional effort to fine-tune the neural network model and explore the precision of the estimated transition model in future research.