# Analysis of dropout learning regarded as ensemble learning

## TL;DR

This paper analyzes dropout learning in deep neural networks as a form of ensemble learning, providing insights into its mechanism and effectiveness in preventing overfitting.

## Contribution

It offers a novel perspective by interpreting dropout as ensemble learning, enhancing understanding of its role in deep learning.

## Key findings

- Dropout can be viewed as an ensemble of thinned networks.
- Ensemble interpretation explains dropout's regularization effect.
- Analysis provides theoretical insights into dropout's success.

## Abstract

Deep learning is the state-of-the-art in fields such as visual object recognition and speech recognition. This learning uses a large number of layers, huge number of units, and connections. Therefore, overfitting is a serious problem. To avoid this problem, dropout learning is proposed. Dropout learning neglects some inputs and hidden units in the learning process with a probability, p, and then, the neglected inputs and hidden units are combined with the learned network to express the final output. We find that the process of combining the neglected hidden units with the learned network can be regarded as ensemble learning, so we analyze dropout learning from this point of view.