Inner Ensemble Networks: Average Ensemble as an Effective Regularizer

Abduallah Mohamed, Muhammed Mohaimin Sadiq, Ehab AlBadawy, Mohamed Elhoseiny, Christian Claudel

Keywords: Regularize, ensemble, variance, reduction

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.

One-sentence Summary: Inner Ensemble Networks adds extra weights during training to improve the model then removes these weights in testing.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2006.08305/code)

Reviewed Version (pdf): https://openreview.net/references/pdf?id=FdlsFYfVF