Domain generalization via model-agnostic learning of semantic features
File(s)
Accepted version
Author(s)
Dou, Q
Coelho De Castro, D
Kamnitsas, K
Glocker, Ben
Type
Conference Paper
Abstract
Generalization capability to unseen domains is crucial for machine learning modelswhen deploying to real-world conditions. We investigate the challenging problemof domain generalization, i.e., training a model on multi-domain source data suchthat it can directly generalize to target domains with unknown statistics. We adopta model-agnostic learning paradigm with gradient-based meta-train and meta-testprocedures to expose the optimization to domain shift. Further, we introducetwo complementary losses which explicitly regularize the semantic structure ofthe feature space. Globally, we align a derived soft confusion matrix to preservegeneral knowledge about inter-class relationships. Locally, we promote domain-independent class-specific cohesion and separation of sample features with ametric-learning component. The effectiveness of our method is demonstrated withnew state-of-the-art results on two common object recognition benchmarks. Ourmethod also shows consistent improvement on a medical image segmentation task.
Date Issued
2019-11-01
Date Acceptance
2019-09-03
Citation
Advances in Neural Information Processing Systems, 2019, 32
ISSN
1049-5258
Publisher
Neural Information Processing Systems Foundation, Inc.
Journal / Book Title
Advances in Neural Information Processing Systems
Volume
32
Copyright Statement
© 2019 Neural Information Processing Systems Foundation, Inc.
Sponsor
Engineering & Physical Science Research Council (E
Commission of the European Communities
Grant Number
EP/R511547/1
H2020 - 757173
Source
Neural Information Processing Systems (NeurIPS)
Subjects
cs.CV
cs.CV
Publication Status
Published online
Start Date
2019-12-08
Finish Date
2019-12-14
Coverage Spatial
Vancouver, Canada