Lu Jiang, Zhengyuan Zhou, Thomas Leung, Li-Jia Li, Li Fei-Fei
Recent deep networks are capable of memorizing the entire data even when the labels are completely random. To overcome the overfitting on corrupted labels, we propose a novel technique of learning another neural network, called MentorNet, to supervise the training of the base deep networks, namely, StudentNet. During training, MentorNet provides a curriculum (sample weighting scheme) for StudentNet to focus on the sample the label of which is probably correct. Unlike the existing curriculum that is usually predefined by human experts, MentorNet learns a data-driven curriculum dynamically with StudentNet. Experimental results demonstrate that our approach can significantly improve the generalization performance of deep networks trained on corrupted training data. Notably, to the best of our knowledge, we achieve the best-published result on WebVision, a large benchmark containing 2.2 million images of real-world noisy labels. The code are at https://github.com/google/mentornet
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| Image Classification | mini WebVision 1.0 | ImageNet Top-1 Accuracy | 63.8 | MentorNet (Inception-ResNet-v2) |
| Image Classification | mini WebVision 1.0 | ImageNet Top-5 Accuracy | 85.8 | MentorNet (Inception-ResNet-v2) |