Benchmark · 2012
A deep convolutional network reduced ImageNet error.
Its benchmark result helped make deep convolutional training a central vision approach.
The result is specific to the ImageNet challenge and its evaluation setup.
NeurIPS · Published 2012
The AlexNet paper trains and evaluates its network on ImageNet images: direct use of the earlier dataset.
ImageNet ClassificationThe AlexNet paper describes training on GPUs using this platform.
ImageNet ClassificationBoth recognise images through layers of local features; the later one ran on much larger data and on GPUs.
Both concern deep convolutional networks for images; the later one removes a limit on depth.
AlexNet in 2012 is the same architecture, enlarged and moved onto a graphics processor.
LeNet-5 and the MNIST set are the direct basis for AlexNet.
The features here are hand-designed; convolutional networks later learn them.
The recipe AlexNet was trained by three years later.
The competition AlexNet won in 2012.
AlexNet in 2012 uses this nonlinearity.
Two ways of proving the same thing in the same year: scale produces features.
The state is given as screen pixels through a convolutional network.
Playing Atari with Deep Reinforcement LearningThe model is built on convolutions, causal and dilated ones.
WaveNet: A generative model for raw audioThe convolution that ruled vision from 2012 turned out to be optional given enough data.
Architectures like AlexNet became available as ready models with weights.
The AlexNet paper (section 4.2) calls dropout a recently introduced technique, applies it with probability 0.5 in the first two fully connected layers and says that without it the network overfits substantially.
ImageNet Classification with Deep Convolutional Neural NetworksR-CNN takes the convolutional network of Krizhevsky and colleagues, pre-trained on ILSVRC 2012, and fine-tunes it on regions for detection.
Rich feature hierarchies for accurate object detection and semantic segmentation (arXiv:1311.2524v1)The paper names the deep convolutional networks of Krizhevsky and colleagues as what made its approach possible.
DeepFace: Closing the Gap to Human-Level Performance in Face Verification