Denoising autoencoder with modulated lateral connections learns invariant representations of natural images
This work addresses a specific challenge in unsupervised learning for computer vision, offering an incremental improvement to autoencoder architectures.
The paper tackled the problem of learning invariant representations in denoising autoencoders by introducing modulated lateral connections between encoder and decoder layers, which improved denoising performance and accelerated the growth of invariance in higher layers.
Suitable lateral connections between encoder and decoder are shown to allow higher layers of a denoising autoencoder (dAE) to focus on invariant representations. In regular autoencoders, detailed information needs to be carried through the highest layers but lateral connections from encoder to decoder relieve this pressure. It is shown that abstract invariant features can be translated to detailed reconstructions when invariant features are allowed to modulate the strength of the lateral connection. Three dAE structures with modulated and additive lateral connections, and without lateral connections were compared in experiments using real-world images. The experiments verify that adding modulated lateral connections to the model 1) improves the accuracy of the probability model for inputs, as measured by denoising performance; 2) results in representations whose degree of invariance grows faster towards the higher layers; and 3) supports the formation of diverse invariant poolings.