Continual Learning Model with Dynamic Adaptation and Flexibility for Incremental Classes
Open Access
- 14 November 2025
- journal article
- research article
- Published by MDPI AG in Applied Sciences
- Vol. 15 (22) , 12118
Abstract
In recent years, deep learning models have achieved highly accurate image classification. However, in scenarios where the number of target classes continues to increase while using a model, it is not very effective to retrain the model by using all data from the beginning each time. Consequently, there has been increasing interest in class-incremental learning, which sequentially updates existing models with newly introduced classes. In particular, Forward Compatible Few-Shot Class-Incremental Learning (FACT) has demonstrated high classification performance by proactively reserving regions within the embedding space to represent knowledge of classes that will be added in the future. In this study, we propose FACT+, which seeks to further enhance the performance of FACT by incorporating two key extensions. First, to account for the semantic diversity (i.e., the breadth or narrowness of meaning) of each class, we introduce variability into each class’s covariance matrix. Second, to address the issue of insufficient learning from incremental data, we implement a mechanism to continuously update the model’s knowledge through the reuse of data collected during deployment. Experimental results show that the proposed method outperforms FACT in terms of classification accuracy and confirm the validity of the embedding space learned by the proposed approach.Keywords
Funding Information
- Japan Society for the Promotion of Science (21H04600, 24H00370)
This publication has 31 references indexed in Scilit:
- Optimizing pre-training via target-aware source data selectionKnowledge-Based Systems, 2025
- Impression evaluation of product images using deep neural networkNeural Computing and Applications, 2025
- Class incremental learning with self-supervised pre-training and prototype learningPattern Recognition, 2024
- Causes of Catastrophic Forgetting in Class-Incremental Semantic Segmentation2023
- Convolutional neural network and its pretrained models for image classification and object detection: A surveyConcurrency and Computation: Practice and Experience, 2021
- Pseudo-rehearsal: Achieving deep reinforcement learning without catastrophic forgettingNeurocomputing, 2021
- Reconciling modern machine-learning practice and the classical bias–variance trade-offProceedings of the National Academy of Sciences, 2019
- On the use of deep learning for blind image quality assessmentSignal, Image and Video Processing, 2017
- A survey of image classification methods and techniques for improving classification performanceInternational Journal of Remote Sensing, 2007
- Catastrophic Interference in Connectionist Networks: The Sequential Learning Problem1989