Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Learning to Optimize Domain Specific Normalization for Domain Generalization

About

We propose a simple but effective multi-source domain generalization technique based on deep neural networks by incorporating optimized normalization layers that are specific to individual domains. Our approach employs multiple normalization methods while learning separate affine parameters per domain. For each domain, the activations are normalized by a weighted average of multiple normalization statistics. The normalization statistics are kept track of separately for each normalization type if necessary. Specifically, we employ batch and instance normalizations in our implementation to identify the best combination of these two normalization methods in each domain. The optimized normalization layers are effective to enhance the generalizability of the learned model. We demonstrate the state-of-the-art accuracy of our algorithm in the standard domain generalization benchmarks, as well as viability to further tasks such as multi-source domain adaptation and domain generalization in the presence of label noise.

Seonguk Seo, Yumin Suh, Dongwan Kim, Geeho Kim, Jongwoo Han, Bohyung Han• 2019

Related benchmarks

TaskDatasetResultRank
Image ClassificationPACS (test)
Average Accuracy86.64
271
Image ClassificationPACS
Overall Average Accuracy85.1
241
Domain GeneralizationPACS
Accuracy86.6
231
Domain GeneralizationPACS (test)
Average Accuracy86.64
225
Image ClassificationOffice-Home (test)
Mean Accuracy62.9
199
Domain GeneralizationPACS (leave-one-domain-out)
Art Accuracy87.04
152
Image ClassificationOffice-Home
Average Accuracy62.9
148
Image ClassificationOfficeHome
Average Accuracy62.9
137
object recognitionPACS (leave-one-domain-out)
Acc (Art painting)87.04
112
Domain GeneralizationOffice-Home (test)
Average Accuracy62.9
106
Showing 10 of 17 rows

Other info

Follow for update