Zhang, Xingzhong and Chuah, Joon Huang and Loo, Chu Kiong and Wermter, Stefan (2024) An Energy Sampling Replay-Based Continual Learning Framework. In: Artificial Neural Networks and Machine Learning-ICANN 2024, Pt II, 17-20 September 2024, Univ Italian Switzerland, Lugano, SWITZERLAND.
Full text not available from this repository.Abstract
Continual Learning represents a significant challenge within the field of computer vision, primarily due to the issue of catastrophic forgetting that arises with sequential learning tasks. Among the array of strategies explored in current continual learning research, replay-based methods have shown notable effectiveness. In this paper, we introduce a novel Energy Sampling Replay-based (ESR) structure for image classification. This framework enhances the selection process of samples for replay by leveraging the energy distribution of the samples, thereby improving the effectiveness of memory samples during the replay phase and increasing accuracy. We have conducted extensive experiments across various continual learning methodologies and datasets. The results demonstrate that our approach effectively mitigates forgetting on CIFAR-10, CIFAR-100 and CIFAR-110 datasets by optimizing the replay strategy.
| Item Type: | Conference or Workshop Item (Paper) |
|---|---|
| Funders: | Universiti Malaya, Department of Artificial Intelligence, Faculty of Computer Science and Information Technology (PPRN001A-2023) |
| Additional Information: | 33rd International Conference on Artificial Neural Networks and Machine Learning (ICANN), Univ Italian Switzerland, Lugano, SWITZERLAND, SEP 17-20, 2024 |
| Uncontrolled Keywords: | Image classification; Continual learning; Catastrophic forgetting; Energy-based sampling |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
| Divisions: | Faculty of Computer Science & Information Technology Faculty of Engineering |
| Depositing User: | Ms. Juhaida Abd Rahim |
| Date Deposited: | 17 Jan 2025 01:32 |
| Last Modified: | 17 Jan 2025 01:32 |
| URI: | http://eprints.um.edu.my/id/eprint/47662 |
Actions (login required)
 |
View Item |
