Reimplementation of Learning to Reweight Examples for Robust Deep Learning

TL;DR

This paper reimplements and evaluates a method for reweighting training examples to improve deep learning robustness, focusing on noisy labels and biases, with applications to skin-cancer detection.

Contribution

It provides a reimplementation of Ren et al.'s approach, verifying its effectiveness on toy and real-world skin-cancer datasets.

Findings

Method improves robustness to noisy labels

Effective in handling dataset biases

Enhances generalization performance

Abstract

Deep neural networks (DNNs) have been used to create models for many complex analysis problems like image recognition and medical diagnosis. DNNs are a popular tool within machine learning due to their ability to model complex patterns and distributions. However, the performance of these networks is highly dependent on the quality of the data used to train the models. Two characteristics of these sets, noisy labels and training set biases, are known to frequently cause poor generalization performance as a result of overfitting to the training set. This paper aims to solve this problem using the approach proposed by Ren et al. (2018) using meta-training and online weight approximation. We will first implement a toy-problem to crudely verify the claims made by the authors of Ren et al. (2018) and then venture into using the approach to solve a real world problem of Skin-cancer detection using an imbalanced image dataset.