KNN-enhanced Deep Learning Against Noisy Labels

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.

[1]  Xiaogang Wang,et al.  Learning from massive noisy labeled data for image classification , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  Samet Oymak,et al.  Gradient Descent with Early Stopping is Provably Robust to Label Noise for Overparameterized Neural Networks , 2019, AISTATS.

[3]  Matthew S. Nokleby,et al.  Learning Deep Networks from Noisy Labels with Dropout Regularization , 2016, 2016 IEEE 16th International Conference on Data Mining (ICDM).

[4]  Richard Nock,et al.  Making Deep Neural Networks Robust to Label Noise: A Loss Correction Approach , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Shih-Chii Liu,et al.  A curriculum learning method for improved noise robustness in automatic speech recognition , 2016, 2017 25th European Signal Processing Conference (EUSIPCO).

[6]  Hamid Parvin,et al.  A Modification on K-Nearest Neighbor Classifier , 2010 .

[7]  Maya R. Gupta,et al.  Deep k-NN for Noisy Labels , 2020, ICML.

[8]  Zhi-Hua Zhou,et al.  On the Resistance of Nearest Neighbor to Random Noisy Labels , 2016 .

[9]  Carla E. Brodley,et al.  Identifying Mislabeled Training Data , 1999, J. Artif. Intell. Res..

[10]  Benjamin Edwards,et al.  Detecting Backdoor Attacks on Deep Neural Networks by Activation Clustering , 2018, SafeAI@AAAI.

[11]  Jason Weston,et al.  Curriculum learning , 2009, ICML '09.

[12]  Patrick D. McDaniel,et al.  Deep k-Nearest Neighbors: Towards Confident, Interpretable and Robust Deep Learning , 2018, ArXiv.

[13]  James Bailey,et al.  Symmetric Cross Entropy for Robust Learning With Noisy Labels , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[14]  Tolga Tasdizen,et al.  Mutual exclusivity loss for semi-supervised deep learning , 2016, 2016 IEEE International Conference on Image Processing (ICIP).

[15]  Jian Sun,et al.  Identity Mappings in Deep Residual Networks , 2016, ECCV.

[16]  Jacob Goldberger,et al.  Training deep neural-networks using a noise adaptation layer , 2016, ICLR.

[17]  Subramanian Ramanathan,et al.  Learning from multiple annotators with varying expertise , 2013, Machine Learning.

[18]  Binqiang Zhao,et al.  O2U-Net: A Simple Noisy Label Detection Approach for Deep Neural Networks , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[19]  Yoshua Bengio,et al.  A Closer Look at Memorization in Deep Networks , 2017, ICML.

[20]  Kun Yi,et al.  Probabilistic End-To-End Noise Correction for Learning With Noisy Labels , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[21]  Dacheng Tao,et al.  Classification with Noisy Labels by Importance Reweighting , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[22]  Harri Valpola,et al.  Weight-averaged consistency targets improve semi-supervised deep learning results , 2017, ArXiv.

[23]  Yannis Avrithis,et al.  Label Propagation for Deep Semi-Supervised Learning , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[24]  Joan Bruna,et al.  Training Convolutional Networks with Noisy Labels , 2014, ICLR 2014.

[25]  Kiyoharu Aizawa,et al.  Joint Optimization Framework for Learning with Noisy Labels , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[26]  Yu-Kun Lai,et al.  Recognition From Web Data: A Progressive Filtering Approach , 2018, IEEE Transactions on Image Processing.

[27]  Dumitru Erhan,et al.  Training Deep Neural Networks on Noisy Labels with Bootstrapping , 2014, ICLR.

[28]  Cheng Soon Ong,et al.  Learning from Corrupted Binary Labels via Class-Probability Estimation , 2015, ICML.

[29]  Jinfeng Yi,et al.  Evaluating the Robustness of Nearest Neighbor Classifiers: A Primal-Dual Perspective , 2019, ArXiv.

[30]  Daphna Weinshall,et al.  On The Power of Curriculum Learning in Training Deep Networks , 2019, ICML.

[31]  Samy Bengio,et al.  Understanding deep learning requires rethinking generalization , 2016, ICLR.

[32]  G LoweDavid,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004 .

[33]  David A. Wagner,et al.  On the Robustness of Deep K-Nearest Neighbors , 2019, 2019 IEEE Security and Privacy Workshops (SPW).

[34]  Antonio Criminisi,et al.  Harvesting Image Databases from the Web , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[35]  Bill Triggs,et al.  Histograms of oriented gradients for human detection , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[36]  King-Sun Fu,et al.  IEEE Transactions on Pattern Analysis and Machine Intelligence Publication Information , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.