PLEASE NOTE CORRECTIONS FOR THIS VIDEO:
The algorithm walkthrough at 5:02 describes a batch size for a given training step, not the overall size of available data!
The hyper-parameters tested in the ablations such as optimizer choice and parameters, learning rate decay, weight decay etc. generalize very well across CIFAR-10, CIFAR-100, SVHN, STL-10, and ImageNet. If you are trying this out for yourself, the hyper-parameter recommendations from the paper should work well without much AutoML / tuning!
This video explains a new algorithm from Google AI for Semi-Supervised Learning! FixMatch uses Consistency Regularization and Pseudo Labeling to achieve about 95% and 89% accuracy when using 250 and 40 labels from CIFAR-10, respectively. Thanks for watching! Please subscribe!
Paper Link (FixMatch): https://arxiv.org/pdf/2001.07685.pdf
Paper used to sort Images according to how representative they are of each class (used for 1 class setting): https://arxiv.org/abs/1910.13427
