Adversarial Adaptive Interpolation in Autoencoders for Dually Regularizing Representation Learning

Abstract

Linear interpolation in the latent space may induce mismatch between the constructed data and the distribution a model was trained on. In this article, we propose an Adversarial Adaptive Interpolation-based AutoEncoder (AdvAI-AE). To constrain the interpolation path on the underlying manifold, an additional interpolation correction module is trained to offset the deviation between the linearly interpolated data points and the statistics of real ones in latent space. Furthermore, we apply prior matching to control the characteristics of the representation. Toward this end, the maximum mean discrepancy-based and adversarial regularizers are incorporated into the model. The synthesized data from random variables are in turn leveraged to regularize the interpolation process. The proposed improvement strategies lead to significant performance gains in downstream classification, clustering, and synthesis tasks on multiple benchmark datasets.