English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In real-world applications, a machine learning model is required to handle anopen-set recognition (OSR), where unknown classes appear during the inference,in addition to a domain shift, where the distribution of data differs betweenthe training and inference phases. Domain generalization (DG) aims to handlethe domain shift situation where the target domain of the inference phase isinaccessible during model training. Open domain generalization (ODG) takes intoaccount both DG and OSR. Domain-Augmented Meta-Learning (DAML) is a methodtargeting ODG but has a complicated learning process. On the other hand,although various DG methods have been proposed, they have not been evaluated inODG situations. This work comprehensively evaluates existing DG methods in ODGand shows that two simple DG methods, CORrelation ALignment (CORAL) and MaximumMean Discrepancy (MMD), are competitive with DAML in several cases. Inaddition, we propose simple extensions of CORAL and MMD by introducing thetechniques used in DAML, such as ensemble learning and Dirichlet mixup dataaugmentation. The experimental evaluation demonstrates that the extended CORALand MMD can perform comparably to DAML with lower computational costs. Thissuggests that the simple DG methods and their simple extensions are strongbaselines for ODG. The code used in the experiments is available athttps://github.com/shiralab/OpenDG-Eval.

Simple Domain Generalization Methods are Strong Baselines for Open Domain Generalization