Self-supervised Latent Space Optimization with Nebula Variational Coding

TL;DR

This paper introduces Nebula Variational Coding, a probabilistic model that optimizes latent space clustering for improved performance across various data types and tasks, using self-supervised metric learning.

Contribution

It proposes a novel variational inference model with nebula anchors to guide latent clustering, enhancing interpretability and adaptability of latent features.

Findings

Effective clustering of latent features across multiple data modalities

Improved performance in classification, segmentation, and reconstruction tasks

Versatile application across text, images, 3D point clouds, and volumetric data

Abstract

Deep learning approaches process data in a layer-by-layer way with intermediate (or latent) features. We aim at designing a general solution to optimize the latent manifolds to improve the performance on classification, segmentation, completion and/or reconstruction through probabilistic models. This paper proposes a variational inference model which leads to a clustered embedding. We introduce additional variables in the latent space, called \textbf{nebula anchors}, that guide the latent variables to form clusters during training. To prevent the anchors from clustering among themselves, we employ the variational constraint that enforces the latent features within an anchor to form a Gaussian distribution, resulting in a generative model we refer as Nebula Variational Coding (NVC). Since each latent feature can be labeled with the closest anchor, we also propose to apply metric learning in a self-supervised way to make the separation between clusters more explicit. As a consequence, the latent variables of our variational coder form clusters which adapt to the generated semantic of the training data, \textit{e.g.} the categorical labels of each sample. We demonstrate experimentally that it can be used within different architectures designed to solve different problems including text sequence, images, 3D point clouds and volumetric data, validating the advantage of our proposed method.