Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385329468> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W4385329468 abstract "Vertical federated learning (VFL) allows multiple clients with misaligned feature spaces to collaboratively accomplish the global model training. Applying VFL to high stakes decision scenarios greatly requires model interpretation for decision reliability and diagnosis. However, the feature discrepancy in VFL raises new issues for model interpretation in distributed setting: one is from the local-global perspective, where the local importance of features is not equal to the global importance; and the other is from the local-local perspective, where information asymmetry among clients causes difficulty in identifying overlapped features. In this work, we propose a new distributed Model Interpretation method for Vertical Federated Learning with feature discrepancy, namely MI-VFL. In particular, to deal with the local-global discrepancy, MI-VFL leverages the law of total probability to adjust the local importance of features and ensures the completeness of the selected features using adversarial game. To handle the local-local discrepancy, MI-VFL builds a federated adversarial learning model to efficiently identify the overlapped features once, rather than performing client-to-client intersections multiple times. We extensively evaluate MI-VFL on six synthetic datasets and five real-world datasets. The evaluation results reveal that MI-VFL can accurately identify the important features, suppress the overlapped features, and thus improve the model performance." @default.
- W4385329468 created "2023-07-29" @default.
- W4385329468 creator A5011060452 @default.
- W4385329468 creator A5050396482 @default.
- W4385329468 creator A5059190563 @default.
- W4385329468 creator A5059753982 @default.
- W4385329468 creator A5068607867 @default.
- W4385329468 date "2023-06-19" @default.
- W4385329468 modified "2023-10-16" @default.
- W4385329468 title "Distributed Model Interpretation for Vertical Federated Learning with Feature Discrepancy" @default.
- W4385329468 cites W1550614472 @default.
- W4385329468 cites W1787224781 @default.
- W4385329468 cites W1968958679 @default.
- W4385329468 cites W2026179596 @default.
- W4385329468 cites W2085988980 @default.
- W4385329468 cites W2104294146 @default.
- W4385329468 cites W2112796928 @default.
- W4385329468 cites W2154053567 @default.
- W4385329468 cites W2516809705 @default.
- W4385329468 cites W2788403449 @default.
- W4385329468 cites W2912213068 @default.
- W4385329468 cites W2945976633 @default.
- W4385329468 cites W2951059495 @default.
- W4385329468 cites W2963233086 @default.
- W4385329468 cites W3103802018 @default.
- W4385329468 cites W3138819813 @default.
- W4385329468 cites W3174440551 @default.
- W4385329468 cites W3209943551 @default.
- W4385329468 cites W4283324687 @default.
- W4385329468 cites W4294106961 @default.
- W4385329468 cites W4296207781 @default.
- W4385329468 doi "https://doi.org/10.1109/iwqos57198.2023.10188708" @default.
- W4385329468 hasPublicationYear "2023" @default.
- W4385329468 type Work @default.
- W4385329468 citedByCount "0" @default.
- W4385329468 crossrefType "proceedings-article" @default.
- W4385329468 hasAuthorship W4385329468A5011060452 @default.
- W4385329468 hasAuthorship W4385329468A5050396482 @default.
- W4385329468 hasAuthorship W4385329468A5059190563 @default.
- W4385329468 hasAuthorship W4385329468A5059753982 @default.
- W4385329468 hasAuthorship W4385329468A5068607867 @default.
- W4385329468 hasConcept C119857082 @default.
- W4385329468 hasConcept C121332964 @default.
- W4385329468 hasConcept C124101348 @default.
- W4385329468 hasConcept C12713177 @default.
- W4385329468 hasConcept C138885662 @default.
- W4385329468 hasConcept C154945302 @default.
- W4385329468 hasConcept C163258240 @default.
- W4385329468 hasConcept C199360897 @default.
- W4385329468 hasConcept C2776401178 @default.
- W4385329468 hasConcept C2992525071 @default.
- W4385329468 hasConcept C41008148 @default.
- W4385329468 hasConcept C41895202 @default.
- W4385329468 hasConcept C43214815 @default.
- W4385329468 hasConcept C527412718 @default.
- W4385329468 hasConcept C59404180 @default.
- W4385329468 hasConcept C62520636 @default.
- W4385329468 hasConceptScore W4385329468C119857082 @default.
- W4385329468 hasConceptScore W4385329468C121332964 @default.
- W4385329468 hasConceptScore W4385329468C124101348 @default.
- W4385329468 hasConceptScore W4385329468C12713177 @default.
- W4385329468 hasConceptScore W4385329468C138885662 @default.
- W4385329468 hasConceptScore W4385329468C154945302 @default.
- W4385329468 hasConceptScore W4385329468C163258240 @default.
- W4385329468 hasConceptScore W4385329468C199360897 @default.
- W4385329468 hasConceptScore W4385329468C2776401178 @default.
- W4385329468 hasConceptScore W4385329468C2992525071 @default.
- W4385329468 hasConceptScore W4385329468C41008148 @default.
- W4385329468 hasConceptScore W4385329468C41895202 @default.
- W4385329468 hasConceptScore W4385329468C43214815 @default.
- W4385329468 hasConceptScore W4385329468C527412718 @default.
- W4385329468 hasConceptScore W4385329468C59404180 @default.
- W4385329468 hasConceptScore W4385329468C62520636 @default.
- W4385329468 hasLocation W43853294681 @default.
- W4385329468 hasOpenAccess W4385329468 @default.
- W4385329468 hasPrimaryLocation W43853294681 @default.
- W4385329468 hasRelatedWork W2546942002 @default.
- W4385329468 hasRelatedWork W2782592381 @default.
- W4385329468 hasRelatedWork W2965871844 @default.
- W4385329468 hasRelatedWork W2970216048 @default.
- W4385329468 hasRelatedWork W3087493185 @default.
- W4385329468 hasRelatedWork W3158586592 @default.
- W4385329468 hasRelatedWork W4206762304 @default.
- W4385329468 hasRelatedWork W4229027123 @default.
- W4385329468 hasRelatedWork W4296012786 @default.
- W4385329468 hasRelatedWork W4361194055 @default.
- W4385329468 isParatext "false" @default.
- W4385329468 isRetracted "false" @default.
- W4385329468 workType "article" @default.