AI summary

This paper explores deep learning architectures and the advantages of unsupervised pre-training, using denoising auto-encoders and two datasets: Shapeset and MNIST. The main result is that pre-training acts as a regularizer, improving generalization and robustness to poor local minima, especially for deeper models.

Abstract

Whereas theoretical work suggests that deep architectures might be more efficient at representing highly-varying functions, training deep architectures was unsuccessful until the recent advent of algorithms based on unsupervised pre-training. Even though these new algorithms have enabled training deep models, many questions remain as to the nature of this difficult learning problem. Answering these questions is important if learning in deep architectures is to be further improved. We attempt to shed some light on these questions through extensive simulations. The experiments confirm and clarify the advantage of unsupervised pre-training. They demonstrate the robustness of the training procedure with respect to the random initialization, the positive effect of pre-training in terms of optimization and its role as a regularizer. We empirically show the influence of pre-training with respect to architecture depth, model capacity, and number of training examples.