KNN-enhanced Deep Learning Against Noisy Labels

Abstract

Supervised learning on Deep Neural Networks (DNNs) is data hungry. Optimizing performance of DNN in the presence of noisy labels has become of paramount importance since collecting a large dataset will usually bring in noisy labels. Inspired by the robustness of K-Nearest Neighbors (KNN) against data noise, in this work, we propose to apply deep KNN for label cleanup. Our approach leverages DNNs for feature extraction and KNN for ground-truth label inference. We iteratively train the neural network and update labels to simultaneously proceed towards higher label recovery rate and better classification performance. Experiment results show that under the same setting, our approach outperforms existing label correction methods and achieves better accuracy on multiple datasets, e.g.,76.78% on Clothing1M dataset.

References in corpus (10)

• Understanding deep learning requires rethinking generalization
• Classification with Noisy Labels by Importance Reweighting
• Training Deep Neural Networks on Noisy Labels with Bootstrapping
• Training Convolutional Networks with Noisy Labels
• Deep k-Nearest Neighbors: Towards Confident, Interpretable and Robust Deep Learning
• Detecting Backdoor Attacks on Deep Neural Networks by Activation Clustering
• Gradient Descent with Early Stopping is Provably Robust to Label Noise for Overparameterized Neural Networks
• On the Resistance of Nearest Neighbor to Random Noisy Labels
• Evaluating the Robustness of Nearest Neighbor Classifiers: A Primal-Dual Perspective
• Deep k-NN for Noisy Labels