How to pronounce“cifar”in English

So let's look at CIFAR-10 as a data set of 10 categories: airplane, automobile, bird, cat, and so on.

is how you have the 95 percentile accuracy on the CIFAR-10 problem.

It's saying if we want to achieve 93% accuracy on ImageNet and 94% on CIFAR10, let's now compete, that's like the requirement,

It's small, it's a small data set: CIFAR-10 means there's 10 categories.

We optimize the hyper-parameters of this algorithm, choose K of 7, seems to work well for this particular CIFAR-10 data set.

Human beings perform at about 94, slightly above 94% accuracy for CIFAR-10.

So training a network that achieves 93% accuracy for 25 bucks, and 94% accuracy for 26 cents on CIFAR10.