Matches in SemOpenAlex for { <https://semopenalex.org/work/W4309130962> ?p ?o ?g. }
Showing items 1 to 74 of
74
with 100 items per page.
- W4309130962 abstract "Data augmentation has been widely used to improve generalization in training deep neural networks. Recent works show that using worst-case transformations or adversarial augmentation strategies can significantly improve the accuracy and robustness. However, due to the non-differentiable properties of image transformations, searching algorithms such as reinforcement learning or evolution strategy have to be applied, which are not computationally practical for large scale problems. In this work, we show that by simply applying consistency training with random data augmentation, state-of-the-art results on domain adaptation (DA) and generalization (DG) can be obtained. To further improve the accuracy and robustness with adversarial examples, we propose a differentiable adversarial data augmentation method based on spatial transformer networks (STN). The combined adversarial and random transformations based method outperforms the state-of-the-art on multiple DA and DG benchmark datasets. Besides, the proposed method shows desirable robustness to corruption, which is also validated on commonly used datasets." @default.
- W4309130962 created "2022-11-23" @default.
- W4309130962 creator A5000375972 @default.
- W4309130962 creator A5000546586 @default.
- W4309130962 creator A5013789507 @default.
- W4309130962 creator A5034078423 @default.
- W4309130962 creator A5049984637 @default.
- W4309130962 creator A5052169513 @default.
- W4309130962 date "2022-11-12" @default.
- W4309130962 modified "2023-10-14" @default.
- W4309130962 title "Adversarial and Random Transformations for Robust Domain Adaptation and Generalization" @default.
- W4309130962 doi "https://doi.org/10.48550/arxiv.2211.06788" @default.
- W4309130962 hasPublicationYear "2022" @default.
- W4309130962 type Work @default.
- W4309130962 citedByCount "0" @default.
- W4309130962 crossrefType "posted-content" @default.
- W4309130962 hasAuthorship W4309130962A5000375972 @default.
- W4309130962 hasAuthorship W4309130962A5000546586 @default.
- W4309130962 hasAuthorship W4309130962A5013789507 @default.
- W4309130962 hasAuthorship W4309130962A5034078423 @default.
- W4309130962 hasAuthorship W4309130962A5049984637 @default.
- W4309130962 hasAuthorship W4309130962A5052169513 @default.
- W4309130962 hasBestOaLocation W43091309621 @default.
- W4309130962 hasConcept C104317684 @default.
- W4309130962 hasConcept C11413529 @default.
- W4309130962 hasConcept C119857082 @default.
- W4309130962 hasConcept C134306372 @default.
- W4309130962 hasConcept C154945302 @default.
- W4309130962 hasConcept C177148314 @default.
- W4309130962 hasConcept C185592680 @default.
- W4309130962 hasConcept C202615002 @default.
- W4309130962 hasConcept C2776434776 @default.
- W4309130962 hasConcept C33923547 @default.
- W4309130962 hasConcept C37736160 @default.
- W4309130962 hasConcept C41008148 @default.
- W4309130962 hasConcept C50644808 @default.
- W4309130962 hasConcept C55493867 @default.
- W4309130962 hasConcept C63479239 @default.
- W4309130962 hasConcept C95623464 @default.
- W4309130962 hasConcept C97541855 @default.
- W4309130962 hasConceptScore W4309130962C104317684 @default.
- W4309130962 hasConceptScore W4309130962C11413529 @default.
- W4309130962 hasConceptScore W4309130962C119857082 @default.
- W4309130962 hasConceptScore W4309130962C134306372 @default.
- W4309130962 hasConceptScore W4309130962C154945302 @default.
- W4309130962 hasConceptScore W4309130962C177148314 @default.
- W4309130962 hasConceptScore W4309130962C185592680 @default.
- W4309130962 hasConceptScore W4309130962C202615002 @default.
- W4309130962 hasConceptScore W4309130962C2776434776 @default.
- W4309130962 hasConceptScore W4309130962C33923547 @default.
- W4309130962 hasConceptScore W4309130962C37736160 @default.
- W4309130962 hasConceptScore W4309130962C41008148 @default.
- W4309130962 hasConceptScore W4309130962C50644808 @default.
- W4309130962 hasConceptScore W4309130962C55493867 @default.
- W4309130962 hasConceptScore W4309130962C63479239 @default.
- W4309130962 hasConceptScore W4309130962C95623464 @default.
- W4309130962 hasConceptScore W4309130962C97541855 @default.
- W4309130962 hasLocation W43091309621 @default.
- W4309130962 hasLocation W43091309622 @default.
- W4309130962 hasOpenAccess W4309130962 @default.
- W4309130962 hasPrimaryLocation W43091309621 @default.
- W4309130962 hasRelatedWork W2773525213 @default.
- W4309130962 hasRelatedWork W2964108292 @default.
- W4309130962 hasRelatedWork W2987815066 @default.
- W4309130962 hasRelatedWork W3100356292 @default.
- W4309130962 hasRelatedWork W3202528422 @default.
- W4309130962 hasRelatedWork W4286663385 @default.
- W4309130962 hasRelatedWork W4288029431 @default.
- W4309130962 hasRelatedWork W4311734044 @default.
- W4309130962 hasRelatedWork W4317552138 @default.
- W4309130962 hasRelatedWork W4319083788 @default.
- W4309130962 isParatext "false" @default.
- W4309130962 isRetracted "false" @default.
- W4309130962 workType "article" @default.