Reconciling Feature-Reuse and Overfitting in DenseNet with Specialized Dropout

TL;DR

This paper introduces a specialized dropout technique for DenseNet CNNs that improves accuracy by addressing overfitting and feature-reuse issues, with effectiveness increasing in deeper models.

Contribution

The paper proposes a novel dropout method tailored for DenseNet's structure, enhancing overfitting prevention without impairing feature-reuse.

Findings

Improved accuracy over vanilla DenseNet and other CNNs

Effectiveness increases with model depth

Dropout location, granularity, and probability are key factors

Abstract

Recently convolutional neural networks (CNNs) achieve great accuracy in visual recognition tasks. DenseNet becomes one of the most popular CNN models due to its effectiveness in feature-reuse. However, like other CNN models, DenseNets also face overfitting problem if not severer. Existing dropout method can be applied but not as effective due to the introduced nonlinear connections. In particular, the property of feature-reuse in DenseNet will be impeded, and the dropout effect will be weakened by the spatial correlation inside feature maps. To address these problems, we craft the design of a specialized dropout method from three aspects, dropout location, dropout granularity, and dropout probability. The insights attained here could potentially be applied as a general approach for boosting the accuracy of other CNN models with similar nonlinear connections. Experimental results show that DenseNets with our specialized dropout method yield better accuracy compared to vanilla DenseNet and state-of-the-art CNN models, and such accuracy boost increases with the model depth.