Y Wu, Y Chen, L Wang, Y Ye, Z Liu… - Proceedings of the …, 2019 - openaccess.thecvf.com
Modern machine learning suffers from catastrophic forgetting when learning new classes incrementally. The performance dramatically degrades due to the missing data of old …
G Lin, H Chu, H Lai - … of the IEEE/CVF Conference on …, 2022 - openaccess.thecvf.com
Plasticity-stability dilemma is a main problem for incremental learning, where plasticity is referring to the ability to learn new knowledge, and stability retains the knowledge of …
Although deep learning approaches have stood out in recent years due to their state-of-the- art results, they continue to suffer from catastrophic forgetting, a dramatic decrease in overall …
R Chen, XY Jing, F Wu, W Zheng, Y Hao - Information Sciences, 2023 - Elsevier
Class incremental learning (CIL) enables deep networks to progressively learn new tasks while remembering previously learned knowledge. A popular design for CIL involves …
JT Zhai, X Liu, AD Bagdanov, K Li… - Proceedings of the …, 2023 - openaccess.thecvf.com
Abstract Class Incremental Learning (CIL) aims to sequentially learn new classes while avoiding catastrophic forgetting of previous knowledge. We propose to use Masked …
E Belouadah, A Popescu - Proceedings of the IEEE/CVF …, 2019 - openaccess.thecvf.com
This paper presents a class incremental learning (IL) method which exploits fine tuning and a dual memory to reduce the negative effect of catastrophic forgetting in image recognition …
E Belouadah, A Popescu - Proceedings of the IEEE/CVF …, 2020 - openaccess.thecvf.com
Incremental learning is useful if an AI agent needs to integrate data from a stream. The problem is non trivial if the agent runs on a limited computational budget and has a bounded …
Neural networks suffer from catastrophic forgetting when sequentially learning tasks phase- by-phase, making them inapplicable in dynamically updated systems. Class-incremental …
X Chen, X Chang - Proceedings of the IEEE/CVF …, 2023 - openaccess.thecvf.com
The rehearsal strategy is widely used to alleviate the catastrophic forgetting problem in class incremental learning (CIL) by preserving limited exemplars from previous tasks. With …