Local Approximations, Real Interpolation and Machine Learning

TL;DR

This paper introduces a new local approximation-based classification algorithm that outperforms shallow neural networks and human predictions on handwritten digit datasets, demonstrating high accuracy on EMNIST.

Contribution

The paper presents a novel classification algorithm connecting local approximations with neural networks and nearest neighbor methods, validated on handwritten digit datasets.

Findings

Misclassification rate of 0.42% on EMNIST

Outperforms shallow ANNs and human predictions

Effective parameter tuning on MNIST

Abstract

We suggest a novel classification algorithm that is based on local approximations and explain its connections with Artificial Neural Networks (ANNs) and Nearest Neighbour classifiers. We illustrate it on the datasets MNIST and EMNIST of images of handwritten digits. We use the dataset MNIST to find parameters of our algorithm and apply it with these parameters to the challenging EMNIST dataset. It is demonstrated that the algorithm misclassifies 0.42% of the images of EMNIST and therefore significantly outperforms predictions by humans and shallow artificial neural networks (ANNs with few hidden layers) that both have more than 1.3% of errors