Few-Shot Class-Incremental Learning

Abstract

The ability to incrementally learn new classes is crucial to the development of real-world artificial intelligence systems. In this paper, we focus on a challenging but practical few-shot class-incremental learning (FSCIL) problem. FSCIL requires CNN models to incrementally learn new classes from very few labelled samples, without forgetting the previously learned ones. To address this problem, we represent the knowledge using a neural gas (NG) network, which can learn and preserve the topology of the feature manifold formed by different classes. On this basis, we propose the TOpology-Preserving knowledge InCrementer (TOPIC) framework. TOPIC mitigates the forgetting of the old classes by stabilizing NG's topology and improves the representation learning for few-shot new classes by growing and adapting NG to new training samples. Comprehensive experimental results demonstrate that our proposed method significantly outperforms other state-of-the-art class-incremental learning methods on CIFAR100, miniImageNet, and CUB200 datasets.

Accepted by CVPR 2020 (oral)

References in corpus (9)

• Distilling the Knowledge in a Neural Network
• Overcoming catastrophic forgetting in neural networks
• Rethinking Atrous Convolution for Semantic Image Segmentation
• Prototypical Networks for Few-shot Learning
• Continual Learning Through Synaptic Intelligence
• Matching Networks for One Shot Learning
• Overcoming catastrophic forgetting with hard attention to the task
• Lifelong Learning with Dynamically Expandable Networks
• Rotate your Networks: Better Weight Consolidation and Less Catastrophic Forgetting

Cited by in corpus (4)

• A Comprehensive Study of Class Incremental Learning Algorithms for Visual Tasks
• Wandering Within a World: Online Contextualized Few-Shot Learning
• Few-shot Continual Learning: a Brain-inspired Approach
• Incremental Embedding Learning via Zero-Shot Translation