Towards continual learning in medical imaging
File(s)
Accepted version
Author(s)
Baweja, C
Glocker, Benjamin
Kamnitsas, K
Type
Conference Paper
Abstract
This work investigates continual learning of two segmentation tasks in brain MRIwith neural networks. To explore in this context the capabilities of current methodsfor countering catastrophic forgetting of the first task when a new one is learned,we investigateelastic weight consolidation[1], a recently proposed method basedon Fisher information, originally evaluated on reinforcement learning of Atarigames. We use it to sequentially learn segmentation of normal brain structures andthen segmentation of white matter lesions. Our findings show this recent methodreduces catastrophic forgetting, while large room for improvement exists in thesechallenging settings for continual learning.
Date Issued
2018-12-08
Date Acceptance
2018-11-05
Citation
2018
Copyright Statement
© 2018 The Author(s)
Sponsor
Commission of the European Communities
Grant Number
H2020 - 757173
Source
Medical imaging meets NIPS
Publication Status
Published
Start Date
2018-12-08
Coverage Spatial
Montreal, Canada