Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387421718> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W4387421718 abstract "We present a novel approach to mitigate bias in facial expression recognition (FER) models. Our method aims to reduce sensitive attribute information such as gender, age, or race, in the embeddings produced by FER models. We employ a kernel mean shrinkage estimator to estimate the kernel mean of the distributions of the embeddings associated with different sensitive attribute groups, such as young and old, in the Hilbert space. Using this estimation, we calculate the maximum mean discrepancy (MMD) distance between the distributions and incorporate it in the classifier loss along with an adversarial loss, which is then minimized through the learning process to improve the distribution alignment. Our method makes sensitive attributes less recognizable for the model, which in turn promotes fairness. Additionally, for the first time, we analyze the notion of attractiveness as an important sensitive attribute in FER models and demonstrate that FER models can indeed exhibit biases towards more attractive faces. To prove the efficacy of our model in reducing bias regarding different sensitive attributes (including the newly proposed attractiveness attribute), we perform several experiments on two widely used datasets, CelebA and RAF-DB. The results in terms of both accuracy and fairness measures outperform the state-of-the-art in most cases, demonstrating the effectiveness of the proposed method." @default.
- W4387421718 created "2023-10-08" @default.
- W4387421718 creator A5039812985 @default.
- W4387421718 creator A5076512643 @default.
- W4387421718 date "2023-10-09" @default.
- W4387421718 modified "2023-10-11" @default.
- W4387421718 title "Toward Fair Facial Expression Recognition with Improved Distribution Alignment" @default.
- W4387421718 cites W1827518550 @default.
- W4387421718 cites W1834627138 @default.
- W4387421718 cites W2100960835 @default.
- W4387421718 cites W2274175710 @default.
- W4387421718 cites W2538496976 @default.
- W4387421718 cites W2560647685 @default.
- W4387421718 cites W2963010714 @default.
- W4387421718 cites W2963116854 @default.
- W4387421718 cites W2963351448 @default.
- W4387421718 cites W2963588172 @default.
- W4387421718 cites W2990751682 @default.
- W4387421718 cites W3005040148 @default.
- W4387421718 cites W3010981289 @default.
- W4387421718 cites W3034700241 @default.
- W4387421718 cites W3048511172 @default.
- W4387421718 cites W3137215626 @default.
- W4387421718 cites W4297678861 @default.
- W4387421718 cites W4312823769 @default.
- W4387421718 cites W4319336174 @default.
- W4387421718 doi "https://doi.org/10.1145/3577190.3614141" @default.
- W4387421718 hasPublicationYear "2023" @default.
- W4387421718 type Work @default.
- W4387421718 citedByCount "0" @default.
- W4387421718 crossrefType "proceedings-article" @default.
- W4387421718 hasAuthorship W4387421718A5039812985 @default.
- W4387421718 hasAuthorship W4387421718A5076512643 @default.
- W4387421718 hasConcept C105795698 @default.
- W4387421718 hasConcept C114614502 @default.
- W4387421718 hasConcept C119857082 @default.
- W4387421718 hasConcept C153180895 @default.
- W4387421718 hasConcept C154945302 @default.
- W4387421718 hasConcept C185429906 @default.
- W4387421718 hasConcept C195704467 @default.
- W4387421718 hasConcept C199360897 @default.
- W4387421718 hasConcept C33923547 @default.
- W4387421718 hasConcept C41008148 @default.
- W4387421718 hasConcept C74193536 @default.
- W4387421718 hasConcept C90559484 @default.
- W4387421718 hasConcept C95623464 @default.
- W4387421718 hasConceptScore W4387421718C105795698 @default.
- W4387421718 hasConceptScore W4387421718C114614502 @default.
- W4387421718 hasConceptScore W4387421718C119857082 @default.
- W4387421718 hasConceptScore W4387421718C153180895 @default.
- W4387421718 hasConceptScore W4387421718C154945302 @default.
- W4387421718 hasConceptScore W4387421718C185429906 @default.
- W4387421718 hasConceptScore W4387421718C195704467 @default.
- W4387421718 hasConceptScore W4387421718C199360897 @default.
- W4387421718 hasConceptScore W4387421718C33923547 @default.
- W4387421718 hasConceptScore W4387421718C41008148 @default.
- W4387421718 hasConceptScore W4387421718C74193536 @default.
- W4387421718 hasConceptScore W4387421718C90559484 @default.
- W4387421718 hasConceptScore W4387421718C95623464 @default.
- W4387421718 hasFunder F4320334593 @default.
- W4387421718 hasLocation W43874217181 @default.
- W4387421718 hasOpenAccess W4387421718 @default.
- W4387421718 hasPrimaryLocation W43874217181 @default.
- W4387421718 hasRelatedWork W2015759683 @default.
- W4387421718 hasRelatedWork W2371687270 @default.
- W4387421718 hasRelatedWork W2642127892 @default.
- W4387421718 hasRelatedWork W3175363083 @default.
- W4387421718 hasRelatedWork W4281634296 @default.
- W4387421718 hasRelatedWork W4285230481 @default.
- W4387421718 hasRelatedWork W4287880334 @default.
- W4387421718 hasRelatedWork W4319161863 @default.
- W4387421718 hasRelatedWork W4366700029 @default.
- W4387421718 hasRelatedWork W4385769873 @default.
- W4387421718 isParatext "false" @default.
- W4387421718 isRetracted "false" @default.
- W4387421718 workType "article" @default.