On Adversarial Mixup Resynthesis
This work addresses representation learning and data synthesis for semi-supervised tasks, but appears incremental as it builds on existing adversarial and mixup techniques.
The paper tackles the problem of combining learned representations in auto-encoders to generate synthetic data that fools an adversarial discriminator, and applies this to semi-supervised learning for class-consistent interpolations, showing quantitative and qualitative evidence of its potential.
In this paper, we explore new approaches to combining information encoded within the learned representations of auto-encoders. We explore models that are capable of combining the attributes of multiple inputs such that a resynthesised output is trained to fool an adversarial discriminator for real versus synthesised data. Furthermore, we explore the use of such an architecture in the context of semi-supervised learning, where we learn a mixing function whose objective is to produce interpolations of hidden states, or masked combinations of latent representations that are consistent with a conditioned class label. We show quantitative and qualitative evidence that such a formulation is an interesting avenue of research.