Task-Agnostic Continual Learning Using Base-Child Classifiers

Page view(s)
45
Checked on Feb 18, 2025
Task-Agnostic Continual Learning Using Base-Child Classifiers
Title:
Task-Agnostic Continual Learning Using Base-Child Classifiers
Journal Title:
2021 IEEE International Conference on Image Processing (ICIP)
Publication Date:
23 August 2021
Citation:
Singh, P. R., Gopalakrishnan, S., ZhongZheng, Q., Suganthan, P. N., Ramasamy, S., & Ambikapathi, A. (2021). Task-Agnostic Continual Learning Using Base-Child Classifiers. 2021 IEEE International Conference on Image Processing (ICIP). doi:10.1109/icip42928.2021.9506504
Abstract:
Continual learning (CL) aims to learn new tasks by forward transfer of information learnt from previous tasks and without forgetting them. In task incremental CL, task information is vital during both strategy development and inference. Providing such partial knowledge about the test sample demands additional complexity and may become intractable, especially when the sample source is ambiguous. In this work, we design a task-agnostic approach that uses base-child hybrid setup to incrementally learn tasks while mitigating forgetting. Multiple base classifiers guided by reference points learn new tasks and this information is distilled via feature space induced sampling strategy. A central child classifier consolidates information across tasks and infers the task identifier automatically. Experimental results on standard datasets show that the proposed approach outperforms the various state-of-the-art regularization and replay CL algorithms in terms of accuracy, by 50% and 7% with homogeneous and heterogeneous tasks, respectively, in task-agnostic scenarios.
License type:
Publisher Copyright
Funding Info:
This research is supported by core funding from: Institute for Infocomm Research (I2R)
Grant Reference no. : SC20/19-128310-CORE
Description:
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
ISSN:
2381-8549
1522-4880
ISBN:
978-1-6654-4115-5
978-1-6654-3102-6
Files uploaded: