Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion

TLDR

The algorithm is a straightforward variation on stacking ordinary autoencoders. The study explores building deep networks by stacking denoising autoencoders trained locally to denoise corrupted inputs. The method stacks denoising autoencoders trained locally to denoise corrupted inputs. On benchmark classification problems, the approach yields significantly lower error than deep belief networks, sometimes surpassing them, and improves SVM performance; qualitative tests show denoising autoencoders learn Gabor‑like edge detectors and larger stroke detectors, demonstrating the value of a denoising criterion for useful higher‑level representations.

Abstract

We explore an original strategy for building deep networks, based on stacking layers of denoising autoencoders which are trained locally to denoise corrupted versions of their inputs. The resulting algorithm is a straightforward variation on the stacking of ordinary autoencoders. It is however shown on a benchmark of classification problems to yield significantly lower classification error, thus bridging the performance gap with deep belief networks (DBN), and in several cases surpassing it. Higher level representations learnt in this purely unsupervised fashion also help boost the performance of subsequent SVM classifiers. Qualitative experiments show that, contrary to ordinary autoencoders, denoising autoencoders are able to learn Gabor-like edge detectors from natural image patches and larger stroke detectors from digit images. This work clearly establishes the value of using a denoising criterion as a tractable unsupervised objective to guide the learning of useful higher level representations.