Self-training with Noisy Student improves ImageNet classification (Paper Explained)

The abundance of data on the internet is vast. Especially unlabeled images are plentiful and can be collected with ease. This model investigates a new method for incorporating unlabeled data into a supervised learning pipeline. First, a teacher model is trained in a supervised fashion. Then, that teacher is used to label the unlabeled data. Next, a larger student model is trained on the combination of all data and achieves better performance than the teacher by itself.

OUTLINE:
0:00 - Intro & Overview
1:05 - Semi-Supervised & Transfer Learning
5:45 - Self-Training & Knowledge Distillation
10:00 - Noisy Student Algorithm Overview
20:20 - Noise Methods
22:30 - Dataset Balancing
25:20 - Results
30:15 - Perturbation Robustness
34:35 - Ablation Studies
39:30 - Conclusion & Comments

Paper: https://arxiv.org/abs/1911.04252
Code: https://github.com/google-research/noisystudent
Models: https://github.com/tensorflow/tpu/tree/master/models/official/efficientnet

Abstract:
We present Noisy Student Training, a semi-supervised learning approach that works well even when labeled data is abundant. Noisy Student Training achieves 88.4% top-1 accuracy on ImageNet, which is 2.0% better than the state-of-the-art model that requires 3.5B weakly labeled Instagram images. On robustness test sets, it improves ImageNet-A top-1 accuracy from 61.0% to 83.7%, reduces ImageNet-C mean corruption error from 45.7 to 28.3, and reduces ImageNet-P mean flip rate from 27.8 to 12.2.
Noisy Student Training extends the idea of self-training and distillation with the use of equal-or-larger student models and noise added to the student during learning. On ImageNet, we first train an EfficientNet model on labeled images and use it as a teacher to generate pseudo labels for 300M unlabeled images. We then train a larger EfficientNet as a student model on the combination of labeled and pseudo labeled images. We iterate this process by putting back the student as the teacher. During the learning of the student, we inject noise such as dropout, stochastic depth, and data augmentation via RandAugment to the student so that the student generalizes better than the teacher. Models are available at this https URL. Code is available at this https URL.

Authors: Qizhe Xie, Minh-Thang Luong, Eduard Hovy, Quoc V. Le

Links:
YouTube: https://www.youtube.com/c/yannickilcher
Twitter: https://twitter.com/ykilcher
Discord: https://discord.gg/4H8xxDF
BitChute: https://www.bitchute.com/channel/yannic-kilcher
Minds: https://www.minds.com/ykilcher
Parler: https://parler.com/profile/YannicKilcher
LinkedIn: https://www.linkedin.com/in/yannic-kilcher-488534136/

If you want to support me, the best thing to do is to share out the content :)

If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this):
SubscribeStar (preferred to Patreon): https://www.subscribestar.com/yannickilcher
Patreon: https://www.patreon.com/yannickilcher
Bitcoin (BTC): bc1q49lsw3q325tr58ygf8sudx2dqfguclvngvy2cq
Ethereum (ETH): 0x7ad3513E3B8f66799f507Aa7874b1B0eBC7F85e2
Litecoin (LTC): LQW2TRyKYetVC8WjFkhpPhtpbDM4Vw7r9m
Monero (XMR): 4ACL8AGrEo5hAir8A9CeVrW8pEauWvnp1WnSDZxW7tziCDLhZAGsgzhRQABDnFy8yuM9fWJDviJPHKRjV4FWt19CJZN9D4n