NAPA-VQ: Neighborhood Aware Prototype Augmentation with Vector Quantization for Continual Learning

TL;DR

NAPA-VQ introduces a novel framework that leverages neighborhood-aware prototype augmentation and vector quantization to improve class discrimination in non-exemplar continual learning, effectively reducing forgetting and class overlap.

Contribution

It proposes a new method combining neighborhood information and vector quantization to enhance class separation without using past exemplars in continual learning.

Findings

Outperforms state-of-the-art NECIL methods on CIFAR-100, TinyImageNet, and ImageNet-Subset.

Achieves 5%, 2%, and 4% higher accuracy respectively.

Reduces forgetting by up to 10%, 3%, and 9%.

Abstract

Catastrophic forgetting; the loss of old knowledge upon acquiring new knowledge, is a pitfall faced by deep neural networks in real-world applications. Many prevailing solutions to this problem rely on storing exemplars (previously encountered data), which may not be feasible in applications with memory limitations or privacy constraints. Therefore, the recent focus has been on Non-Exemplar based Class Incremental Learning (NECIL) where a model incrementally learns about new classes without using any past exemplars. However, due to the lack of old data, NECIL methods struggle to discriminate between old and new classes causing their feature representations to overlap. We propose NAPA-VQ: Neighborhood Aware Prototype Augmentation with Vector Quantization, a framework that reduces this class overlap in NECIL. We draw inspiration from Neural Gas to learn the topological relationships in the feature space, identifying the neighboring classes that are most likely to get confused with each other. This neighborhood information is utilized to enforce strong separation between the neighboring classes as well as to generate old class representative prototypes that can better aid in obtaining a discriminative decision boundary between old and new classes. Our comprehensive experiments on CIFAR-100, TinyImageNet, and ImageNet-Subset demonstrate that NAPA-VQ outperforms the State-of-the-art NECIL methods by an average improvement of 5%, 2%, and 4% in accuracy and 10%, 3%, and 9% in forgetting respectively. Our code can be found in https://github.com/TamashaM/NAPA-VQ.git.