Multivariate Temporal Autoencoder for Predictive Reconstruction of Deep Sequences

Abstract

Time series sequence prediction and modelling has proven to be a challenging endeavor in real world datasets. Two key issues are the multi-dimensionality of data and the interaction of independent dimensions forming a latent output signal, as well as the representation of multi-dimensional temporal data inside of a predictive model. This paper proposes a multi-branch deep neural network approach to tackling the aforementioned problems by modelling a latent state vector representation of data windows through the use of a recurrent autoencoder branch and subsequently feeding the trained latent vector representation into a predictor branch of the model. This model is henceforth referred to as Multivariate Temporal Autoencoder (MvTAe). The framework in this paper utilizes a synthetic multivariate temporal dataset which contains dimensions that combine to create a hidden output target.