MaxDropout: Deep Neural Network Regularization Based on Maximum Output Values

TL;DR

MaxDropout is a novel regularization technique for deep neural networks that enhances sparsity by shutting off the most active neurons, leading to improved or comparable accuracy in image classification tasks.

Contribution

Introduces MaxDropout, a new regularizer that removes the most active neurons to promote sparsity and improve neural network performance.

Findings

MaxDropout improves accuracy over Dropout in neural networks.

MaxDropout achieves comparable results to Cutout and RandomErasing.

Replacing Dropout with MaxDropout enhances neural network performance.

Abstract

Different techniques have emerged in the deep learning scenario, such as Convolutional Neural Networks, Deep Belief Networks, and Long Short-Term Memory Networks, to cite a few. In lockstep, regularization methods, which aim to prevent overfitting by penalizing the weight connections, or turning off some units, have been widely studied either. In this paper, we present a novel approach called MaxDropout, a regularizer for deep neural network models that works in a supervised fashion by removing (shutting off) the prominent neurons (i.e., most active) in each hidden layer. The model forces fewer activated units to learn more representative information, thus providing sparsity. Regarding the experiments, we show that it is possible to improve existing neural networks and provide better results in neural networks when Dropout is replaced by MaxDropout. The proposed method was evaluated in image classification, achieving comparable results to existing regularizers, such as Cutout and RandomErasing, also improving the accuracy of neural networks that uses Dropout by replacing the existing layer by MaxDropout.