Partitioned Memory Storage Inspired Few-Shot Class-Incremental Learning
Open Access
Abstract
Current mainstream deep learning techniques exhibit an over-reliance on extensive training data and a lack of adaptability to the dynamic world, marking a considerable disparity from human intelligence. To bridge this gap, Few-Shot Class-Incremental Learning(FSCIL) has emerged, focusing on continuous learning of new categories with limited samples without forgetting old knowledge. Existing FSCIL studies typically use a single model to learn knowledge across all sessions, inevitably leading to the stability–plasticity dilemma. Unlike machines that usually consolidate all categories into a single parameter space, cortical memory organization suggests that different types of knowledge can be distributed and organized across specialized cortical regions. Inspired by this organization principle, our paper aims to develop a method that learns independent models for each session. It can inherently prevent catastrophic forgetting. During the testing stage, our method integrates Uncertainty Quantification (UQ) for model deployment. Our method provides a fresh viewpoint for FSCIL and demonstrates the state-of-the-art performance on CIFAR-100 and mini-ImageNet datasets.Keywords
Funding Information
- Fundamental Research Funds for the Central Universities (B250201270)
- Outstanding Youth Project of the Hunan Provincial Department of Education (24B0865)
- National Natural Science Foundation of China (62306323)
- Science and Technology Program of Hunan Province (2024RC3117)
This publication has 23 references indexed in Scilit:
- Few-Shot Class-Incremental Learning via Entropy-Regularized Data-Free Replay2022
- FOSTER: Feature Boosting and Compression for Class-Incremental Learning2022
- Class-Incremental Learning: Survey and Performance Evaluation on Image ClassificationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2022
- Biological underpinnings for lifelong learning machinesNature Machine Intelligence, 2022
- A continual learning survey: Defying forgetting in classification tasksIEEE Transactions on Pattern Analysis and Machine Intelligence, 2021
- Generalizing from a Few ExamplesACM Computing Surveys, 2020
- End-to-End Incremental Learning2018
- PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruning2018
- iCaRL: Incremental Classifier and Representation Learning2017
- Deep Residual Learning for Image Recognition2016