Leveraging Hierarchical Taxonomies in Prompt-based Continual Learning

TL;DR

This paper introduces a hierarchical taxonomy-based regularization method for prompt-based continual learning, inspired by human hierarchical organization, to reduce catastrophic forgetting and improve model performance.

Contribution

It proposes a novel hierarchical regularization approach leveraging class relationships and pretrained model insights to enhance continual learning.

Findings

Significant performance improvements over state-of-the-art models.

Effective identification of confusing class groups using hierarchical structures.

Enhanced focus on challenging knowledge areas through new regularization loss.

Abstract

Humans perceive the world as a series of sequential events, which can be hierarchically organized with different levels of abstraction based on conceptual knowledge. Drawing inspiration from human learning behaviors, this work proposes a novel approach to mitigate catastrophic forgetting in Prompt-based Continual Learning models by exploiting the relationships between continuously emerging class data. We find that applying human habits of organizing and connecting information can serve as an efficient strategy when training deep learning models. Specifically, by building a hierarchical tree structure based on the expanding set of labels, we gain fresh insights into the data, identifying groups of similar classes could easily cause confusion. Additionally, we delve deeper into the hidden connections between classes by exploring the original pretrained model's behavior through an optimal transport-based approach. From these insights, we propose a novel regularization loss function that encourages models to focus more on challenging knowledge areas, thereby enhancing overall performance. Experimentally, our method demonstrated significant superiority over the most robust state-of-the-art models on various benchmarks.