An Optimal Transport-driven Approach for Cultivating Latent Space in Online Incremental Learning

TL;DR

This paper introduces MMOT, an online Mixture Model framework based on Optimal Transport, to improve latent space representation and class separation in online incremental learning with distributional shifts.

Contribution

It proposes a novel MMOT framework with incremental centroid updates and a Dynamic Preservation strategy to enhance class modeling and prevent catastrophic forgetting.

Findings

MMOT outperforms existing methods on benchmark datasets.

The approach better captures complex data streams with multimodal distributions.

Dynamic Preservation maintains class separability over time.

Abstract

In online incremental learning, data continuously arrives with substantial distributional shifts, creating a significant challenge because previous samples have limited replay value when learning a new task. Prior research has typically relied on either a single adaptive centroid or multiple fixed centroids to represent each class in the latent space. However, such methods struggle when class data streams are inherently multimodal and require continual centroid updates. To overcome this, we introduce an online Mixture Model learning framework grounded in Optimal Transport theory (MMOT), where centroids evolve incrementally with new data. This approach offers two main advantages: (i) it provides a more precise characterization of complex data streams, and (ii) it enables improved class similarity estimation for unseen samples during inference through MMOT-derived centroids. Furthermore, to strengthen representation learning and mitigate catastrophic forgetting, we design a Dynamic Preservation strategy that regulates the latent space and maintains class separability over time. Experimental evaluations on benchmark datasets confirm the superior effectiveness of our proposed method.