Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387211568> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W4387211568 endingPage "735" @default.
- W4387211568 startingPage "725" @default.
- W4387211568 abstract "Adversarial training has been demonstrated to be one of the most effective approaches to training deep neural networks that are robust to malicious perturbations. Research on effectively applying it to produce robust 3D medical image segmentation models is ongoing. While few empirical studies have been done in this area, developing effective adversarial training methods for complex segmentation models and high-volume 3D examples is challenging and requires theoretical support. In this paper, we consider the robustness of 3D segmentation tasks from a PAC-Bayes generalisation perceptive and show that reducing the trained models’ Lipschitz constant benefits the models’ robustness performance. Demonstrating by empirical investigation, we show that adjusting the adversarial iteration can help to reduce the model’s Lipschitz constant, enabling a self-adaptive adversarial training strategy. Empirical studies on the medical segmentation decathlon dataset have been done to demonstrate the efficiency of the proposed adversarial training method. Our implementation is available at https://github.com/TrustAI/SEAT ." @default.
- W4387211568 created "2023-10-01" @default.
- W4387211568 creator A5012939991 @default.
- W4387211568 creator A5046599212 @default.
- W4387211568 creator A5074225885 @default.
- W4387211568 creator A5081956313 @default.
- W4387211568 date "2023-01-01" @default.
- W4387211568 modified "2023-10-01" @default.
- W4387211568 title "Self-adaptive Adversarial Training for Robust Medical Segmentation" @default.
- W4387211568 cites W1901129140 @default.
- W4387211568 cites W2029029543 @default.
- W4387211568 cites W2604505099 @default.
- W4387211568 cites W2963026800 @default.
- W4387211568 cites W2964097310 @default.
- W4387211568 cites W2976398475 @default.
- W4387211568 cites W3011708665 @default.
- W4387211568 cites W3021182036 @default.
- W4387211568 cites W3027789868 @default.
- W4387211568 cites W3030790048 @default.
- W4387211568 cites W3036286896 @default.
- W4387211568 cites W3090005963 @default.
- W4387211568 cites W3112701542 @default.
- W4387211568 cites W3172681723 @default.
- W4387211568 cites W3201839864 @default.
- W4387211568 cites W3203275379 @default.
- W4387211568 cites W3204235424 @default.
- W4387211568 cites W4221114735 @default.
- W4387211568 cites W4283732098 @default.
- W4387211568 cites W4296123053 @default.
- W4387211568 cites W4312248169 @default.
- W4387211568 cites W4353044157 @default.
- W4387211568 cites W4367362824 @default.
- W4387211568 doi "https://doi.org/10.1007/978-3-031-43898-1_69" @default.
- W4387211568 hasPublicationYear "2023" @default.
- W4387211568 type Work @default.
- W4387211568 citedByCount "0" @default.
- W4387211568 crossrefType "book-chapter" @default.
- W4387211568 hasAuthorship W4387211568A5012939991 @default.
- W4387211568 hasAuthorship W4387211568A5046599212 @default.
- W4387211568 hasAuthorship W4387211568A5074225885 @default.
- W4387211568 hasAuthorship W4387211568A5081956313 @default.
- W4387211568 hasConcept C104317684 @default.
- W4387211568 hasConcept C108583219 @default.
- W4387211568 hasConcept C119857082 @default.
- W4387211568 hasConcept C134306372 @default.
- W4387211568 hasConcept C154945302 @default.
- W4387211568 hasConcept C185592680 @default.
- W4387211568 hasConcept C22324862 @default.
- W4387211568 hasConcept C2984842247 @default.
- W4387211568 hasConcept C33923547 @default.
- W4387211568 hasConcept C37736160 @default.
- W4387211568 hasConcept C41008148 @default.
- W4387211568 hasConcept C55493867 @default.
- W4387211568 hasConcept C63479239 @default.
- W4387211568 hasConcept C89600930 @default.
- W4387211568 hasConceptScore W4387211568C104317684 @default.
- W4387211568 hasConceptScore W4387211568C108583219 @default.
- W4387211568 hasConceptScore W4387211568C119857082 @default.
- W4387211568 hasConceptScore W4387211568C134306372 @default.
- W4387211568 hasConceptScore W4387211568C154945302 @default.
- W4387211568 hasConceptScore W4387211568C185592680 @default.
- W4387211568 hasConceptScore W4387211568C22324862 @default.
- W4387211568 hasConceptScore W4387211568C2984842247 @default.
- W4387211568 hasConceptScore W4387211568C33923547 @default.
- W4387211568 hasConceptScore W4387211568C37736160 @default.
- W4387211568 hasConceptScore W4387211568C41008148 @default.
- W4387211568 hasConceptScore W4387211568C55493867 @default.
- W4387211568 hasConceptScore W4387211568C63479239 @default.
- W4387211568 hasConceptScore W4387211568C89600930 @default.
- W4387211568 hasLocation W43872115681 @default.
- W4387211568 hasOpenAccess W4387211568 @default.
- W4387211568 hasPrimaryLocation W43872115681 @default.
- W4387211568 hasRelatedWork W2950183588 @default.
- W4387211568 hasRelatedWork W3193857078 @default.
- W4387211568 hasRelatedWork W3208304128 @default.
- W4387211568 hasRelatedWork W3208723233 @default.
- W4387211568 hasRelatedWork W4311734044 @default.
- W4387211568 hasRelatedWork W4320076403 @default.
- W4387211568 hasRelatedWork W4322759769 @default.
- W4387211568 hasRelatedWork W4379255972 @default.
- W4387211568 hasRelatedWork W4383955378 @default.
- W4387211568 hasRelatedWork W4286890323 @default.
- W4387211568 isParatext "false" @default.
- W4387211568 isRetracted "false" @default.
- W4387211568 workType "book-chapter" @default.