Impact Statement: Autoencoder is a popular data-driven modeling technology in deep learning. It can deal with the nonlinear relationships among process variables, and has a powerful feature extraction ...

This paper proposes a complex recurrent variational autoencoder (VAE) framework, for modeling time series data, particularly speech signals. First, to account for the temporal structure of speech ...

A prototype system that uses a CNN encoder and edge-based features to retrieve visually similar fashion items from the Fashion MNIST dataset using cosine similarity.

This is an autoencoder with cylic loss and coding parsing loss for image compression and reconstruction. Network backbone is simple 3-layer fully conv (encoder) and symmetrical for decoder. Finally it ...