In this work we seek to bridge the concepts of topographic organization and equiv-ariance in neural networks. To accomplish this, we introduce the Topographic V AE: a novel method for efficiently training deep generative models with topographically organized latent variables.

On the other hand, representation learning at part level has received significantly less attention.

Our method jointly optimizes a latent-space model and policy to be self-consistent, such that the policy avoids states where the model is inaccurate.

Most existing PLL approaches assume that the incorrect labels in each training example are randomly picked as the candidate labels. However, this assumption is not realistic since the candidate labels are always instance-dependent. In this paper, we consider instance-dependent PLL and assume that each example is associated with a latent label distribution constituted by the real number of each label, representing the degree to each label describing the feature. The incorrect label with a high degree is more likely to be annotated as the candidate label.