Google Brain & CMU Semi-Supervised ‘Noisy Student’ Achieves 88.4% Top-1 Accuracy on ImageNet

Google Brain & CMU Semi-Supervised ‘Noisy Student’ Achieves 88.4% Top-1 Accuracy on ImageNet

The proposed method’s 88.4 percent accuracy on ImageNet is 2.0 percent better than the SOTA model.

5 mentions: @Synced_Global@Synced_Global
Keywords: imagenet
Date: 2020/02/13 16:56

Referring Tweets

@Synced_Global 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. #DeepLearning #machinelearning #AI t.co/xrqJmz3jsi
@Synced_Global Researchers from @GoogleAI Brain Team and @CarnegieMellon have released models trained with a semi-supervised learning method called “Noisy Student” that achieve 88.4 percent top-1 accuracy on ImageNet. #SOTA #DeepLearning #MachineLearning t.co/xrqJmyLIAK

Related Entries

Read more NLP's ImageNet moment has arrived
0 users, 0 mentions 2018/07/09 18:22
Read more GitHub - Tencent/tencent-ml-images: Largest multi-label image database; ResNet-101 model; 80.73% top...
0 users, 12 mentions 2018/12/02 15:01
Read more Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour – Facebook Research
0 users, 0 mentions 2018/04/22 03:40
Read more ImageNet Large Scale Visual Recognition Competition 2012 (ILSVRC2012)
0 users, 1 mentions 2019/01/19 06:45
Read more Training Imagenet in 3 hours for $25; and CIFAR10 for $0.26 · fast.ai
0 users, 0 mentions 2018/05/04 16:46