Multi-column Deep Neural Networks for Image Classification

TL;DR

This paper introduces multi-column deep neural networks inspired by biological vision, achieving near-human performance on MNIST and surpassing humans on traffic sign recognition, with state-of-the-art results on various benchmarks.

Contribution

The paper presents a biologically inspired deep neural network architecture with multiple columns and winner-take-all neurons, achieving superior image classification performance.

Findings

Achieved near-human accuracy on MNIST handwritten digit recognition.

Outperformed humans by a factor of two on traffic sign recognition.

Set new state-of-the-art results on multiple image classification benchmarks.

Abstract

Traditional methods of computer vision and machine learning cannot match human performance on tasks such as the recognition of handwritten digits or traffic signs. Our biologically plausible deep artificial neural network architectures can. Small (often minimal) receptive fields of convolutional winner-take-all neurons yield large network depth, resulting in roughly as many sparsely connected neural layers as found in mammals between retina and visual cortex. Only winner neurons are trained. Several deep neural columns become experts on inputs preprocessed in different ways; their predictions are averaged. Graphics cards allow for fast training. On the very competitive MNIST handwriting benchmark, our method is the first to achieve near-human performance. On a traffic sign recognition benchmark it outperforms humans by a factor of two. We also improve the state-of-the-art on a plethora of common image classification benchmarks.