High-Performance Neural Networks for Visual Object Classification

TL;DR

This paper introduces a fast, GPU-based implementation of deep convolutional neural networks that learn features in a supervised manner, achieving state-of-the-art results on multiple object classification benchmarks with rapid training times.

Contribution

It presents a fully parameterizable GPU implementation of CNNs with learned features, demonstrating superior performance and fast training on standard benchmarks.

Findings

Achieved 2.53% error on NORB dataset

Achieved 19.51% error on CIFAR10 dataset

Achieved 0.35% error on MNIST dataset

Abstract

We present a fast, fully parameterizable GPU implementation of Convolutional Neural Network variants. Our feature extractors are neither carefully designed nor pre-wired, but rather learned in a supervised way. Our deep hierarchical architectures achieve the best published results on benchmarks for object classification (NORB, CIFAR10) and handwritten digit recognition (MNIST), with error rates of 2.53%, 19.51%, 0.35%, respectively. Deep nets trained by simple back-propagation perform better than more shallow ones. Learning is surprisingly rapid. NORB is completely trained within five epochs. Test error rates on MNIST drop to 2.42%, 0.97% and 0.48% after 1, 3 and 17 epochs, respectively.