.png){kind=icon}
.png){kind=icon}
Return
Continual Learning With Unknown Task Boundary
.png){kind=icon}
PRE
AI
DOI：10.1109/TNNLS.2024.3412934.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Most existing studies on continual learning (CL) consider the task-based setting, where task boundaries are known to learners during training. However, they may be impractical for real-world problems, where new tasks arrive with unnotified distribution shifts. In this article, we introduce a new boundary-unknown continual learning scenario called continuum incremental learning (CoIL), where the incremental unit may be a concatenation of several tasks or a subset of one task. To identify task boundaries, we design a continual out-of-distribution (OOD) detection method based on softmax probabilities, which can detect OOD samples for the latest learned task. Then, we incorporate it with continual learning approaches to solve the CoIL problem. Furthermore, we investigate the more challenging task-reappear setting and propose a method named continual learning with unknown task boundary (CLUTaB). CLUTaB first adopts in-distribution detection and OOD loss to determine whether a set of data is sampled from any learned distribution. Then, a two-step inference technique is designed to improve the continual learning performance. Experiments show that our methods work well with existing continual learning approaches and achieve good performance on CIFAR-100 and mini-ImageNet datasets.
Keywords:
Task analysis
Continuing education
Coils
Training
Training data
Software
Sharks
Continual learning (CL)
continuum incremental learning (CoIL)
deep learning
out-of-distribution (OOD) detection
Journal
IF:
8.9
Papers:
7.5K
Citations:
7.2W