Probabilistic End-To-End Noise Correction for Learning With Noisy Labels

Deep learning has achieved excellent performance in various computer vision tasks, but requires a lot of training examples with clean labels. It is easy to collect a dataset with noisy labels, but such noise makes networks overfit seriously and accuracies drop dramatically. To address this problem, we propose an end-to-end framework called PENCIL, which can update both network parameters and label estimations as label distributions. PENCIL is independent of the backbone network structure and does not need an auxiliary clean dataset or prior information about noise, thus it is more general and robust than existing methods and is easy to apply. PENCIL outperformed previous state-of-the-art methods by large margins on both synthetic and real-world datasets with different noise types and noise rates. Experiments show that PENCIL is robust on clean datasets, too.

[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]  Jianxin Wu,et al.  Deep Label Distribution Learning With Label Ambiguity , 2016, IEEE Transactions on Image Processing.

[3]  David J. Kriegman,et al.  Learning Concept Embeddings with Combined Human-Machine Expertise , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[4]  Dana Angluin,et al.  Learning from noisy examples , 1988, Machine Learning.

[5]  Arash Vahdat,et al.  Toward Robustness against Label Noise in Training Deep Discriminative Neural Networks , 2017, NIPS.

[6]  N. Mati,et al.  Discovering Informative Patterns and Data Cleaning , 1996 .

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

[8]  Aritra Ghosh,et al.  Robust Loss Functions under Label Noise for Deep Neural Networks , 2017, AAAI.

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

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

[11]  Lars Kai Hansen,et al.  Design of robust neural network classifiers , 1998, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181).

[12]  Mert R. Sabuncu,et al.  Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labels , 2018, NeurIPS.

[13]  Pietro Perona,et al.  Learning Object Categories From Internet Image Searches , 2010, Proceedings of the IEEE.

[14]  J. R. Quinlan.  Induction of decision trees , 2004, Machine Learning.

[15]  Fei-Fei Li,et al.  ImageNet: A large-scale hierarchical image database , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[16]  James Bailey,et al.  Dimensionality-Driven Learning with Noisy Labels , 2018, ICML.

[17]  Chen Sun,et al.  Revisiting Unreasonable Effectiveness of Data in Deep Learning Era , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[18]  Alex Krizhevsky,et al.  Learning Multiple Layers of Features from Tiny Images , 2009 .

[19]  Jun Sun,et al.  Deep Learning From Noisy Image Labels With Quality Embedding , 2017, IEEE Transactions on Image Processing.

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

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

[22]  Pietro Perona,et al.  Caltech-UCSD Birds 200 , 2010 .

[23]  Le Song,et al.  Iterative Learning with Open-set Noisy Labels , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

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

[25]  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).

[26]  Jonathan Krause,et al.  The Unreasonable Effectiveness of Noisy Data for Fine-Grained Recognition , 2015, ECCV.

[27]  Nir Shavit,et al.  Deep Learning is Robust to Massive Label Noise , 2017, ArXiv.

[28]  M. Verleysen,et al.  Classification in the Presence of Label Noise: A Survey , 2014, IEEE Transactions on Neural Networks and Learning Systems.

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

[30]  Lei Zhang,et al.  CleanNet: Transfer Learning for Scalable Image Classifier Training with Label Noise , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[31]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).