On Diversity in Discriminative Neural Networks

TL;DR

This paper introduces a neural network architecture leveraging diversity principles, achieving state-of-the-art results in self-supervised and semi-supervised learning on MNIST and CIFAR-10 datasets.

Contribution

It proposes a novel neural network architecture that incorporates various diversity principles, some original, to improve learning performance.

Findings

Achieved 99.57% accuracy in self-supervised MNIST

Achieved 94.21% accuracy in semi-supervised CIFAR-10 with 25 labels per class

Demonstrated the effectiveness of diversity principles in neural network design

Abstract

Diversity is a concept of prime importance in almost all disciplines based on information processing. In telecommunications, for example, spatial, temporal, and frequency diversity, as well as redundant coding, are fundamental concepts that have enabled the design of extremely efficient systems. In machine learning, in particular with neural networks, diversity is not always a concept that is emphasized or at least clearly identified. This paper proposes a neural network architecture that builds upon various diversity principles, some of them already known, others more original. Our architecture obtains remarkable results, with a record self-supervised learning accuracy of 99. 57% in MNIST, and a top tier promising semi-supervised learning accuracy of 94.21% in CIFAR-10 using only 25 labels per class.