Inner Ensemble Networks: Average Ensemble as an Effective Regularizer

TL;DR

Inner Ensemble Networks (IENs) are a novel regularization method that reduces neural network variance through internal ensemble parameters during training, leading to improved performance without increasing model complexity.

Contribution

This paper introduces IENs, a new approach that reduces variance within neural networks using ensemble parameters, outperforming existing methods like dropout and maxout.

Findings

IENs reduce network variance by a factor of 1/m^{L-1}.

IENs decrease error rates by 1.7% to 17.3%.

IENs are preferred by Neural Architecture Search methods.

Abstract

We introduce Inner Ensemble Networks (IENs) which reduce the variance within the neural network itself without an increase in the model complexity. IENs utilize ensemble parameters during the training phase to reduce the network variance. While in the testing phase, these parameters are removed without a change in the enhanced performance. IENs reduce the variance of an ordinary deep model by a factor of $1/m^{L-1}$, where $m$ is the number of inner ensembles and $L$ is the depth of the model. Also, we show empirically and theoretically that IENs lead to a greater variance reduction in comparison with other similar approaches such as dropout and maxout. Our results show a decrease of error rates between 1.7\% and 17.3\% in comparison with an ordinary deep model. We also show that IEN was preferred by Neural Architecture Search (NAS) methods over prior approaches. Code is available at https://github.com/abduallahmohamed/inner_ensemble_nets.