Multi-pretrained Deep Neural Network
This work addresses the choice of pretraining methods for deep learning practitioners, but it is incremental as it builds on existing models without introducing new techniques.
The paper compared Deep Belief Network (DBN) and Stacked Denoising Autoencoder (SDA) as pretraining models for deep neural networks, finding that DBN provides a better initial model but SDA leads to better final performance after fine-tuning, especially when used in a multi-pretraining approach.
Pretraining is widely used in deep neutral network and one of the most famous pretraining models is Deep Belief Network (DBN). The optimization formulas are different during the pretraining process for different pretraining models. In this paper, we pretrained deep neutral network by different pretraining models and hence investigated the difference between DBN and Stacked Denoising Autoencoder (SDA) when used as pretraining model. The experimental results show that DBN get a better initial model. However the model converges to a relatively worse model after the finetuning process. Yet after pretrained by SDA for the second time the model converges to a better model if finetuned.