ImageNet Classification with Deep Convolutional Neural Networks

This paper discussed the problem of image classification on a subset of the ImageNet data set, the LSRVC-2010. It introduces AlexNet, a deep Convolutional Neural Network architecture, which won the 2012 ImageNet competition, significantly advancing deep learning.

The paper discusses the Model Architecture, the training process, and the results this model achieves. The model has 8 layers, 5 convolutional and 3 connected layers, and uses the ReLU activation function that helps speed the training process. AlexNet incorporates data augmentation techniques like random cropping and horizontal flipping to solve the problem of overfitting. AlexNet achieved top-1 and top-5 error rates of 37.5% and 17.0% which was significantly better than other models at that time, highlighting the potential of deep learning in large-scale image classification