FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3162922192> ?p ?o ?g. }

• W3162922192 endingPage "2307" @default.
• W3162922192 startingPage "2288" @default.
• W3162922192 abstract "Abstract Deep learning has catalysed progress in tasks such as face recognition and analysis, leading to a quick integration of technological solutions in multiple layers of our society. While such systems have proven to be accurate by standard evaluation metrics and benchmarks, a surge of work has recently exposed the demographic bias that such algorithms exhibit–highlighting that accuracy does not entail fairness . Clearly, deploying biased systems under real-world settings can have grave consequences for affected populations. Indeed, learning methods are prone to inheriting, or even amplifying the bias present in a training set, manifested by uneven representation across demographic groups. In facial datasets, this particularly relates to attributes such as skin tone , gender , and age . In this work, we address the problem of mitigating bias in facial datasets by data augmentation. We propose a multi-attribute framework that can successfully transfer complex, multi-scale facial patterns even if these belong to underrepresented groups in the training set. This is achieved by relaxing the rigid dependence on a single attribute label, and further introducing a tensor-based mixing structure that captures multiplicative interactions between attributes in a multilinear fashion. We evaluate our method with an extensive set of qualitative and quantitative experiments on several datasets, with rigorous comparisons to state-of-the-art methods. We find that the proposed framework can successfully mitigate dataset bias, as evinced by extensive evaluations on established diversity metrics, while significantly improving fairness metrics such as equality of opportunity." @default.
• W3162922192 created "2021-05-24" @default.
• W3162922192 creator A5016033078 @default.
• W3162922192 creator A5017437981 @default.
• W3162922192 creator A5046779014 @default.
• W3162922192 creator A5050734738 @default.
• W3162922192 creator A5074117139 @default.
• W3162922192 date "2021-05-15" @default.
• W3162922192 modified "2023-10-01" @default.
• W3162922192 title "Mitigating Demographic Bias in Facial Datasets with Style-Based Multi-attribute Transfer" @default.
• W3162922192 cites W1834627138 @default.
• W3162922192 cites W1967022358 @default.
• W3162922192 cites W2000215628 @default.
• W3162922192 cites W2024165284 @default.
• W3162922192 cites W2105026179 @default.
• W3162922192 cites W2115651492 @default.
• W3162922192 cites W2118664399 @default.
• W3162922192 cites W2134251598 @default.
• W3162922192 cites W2194775991 @default.
• W3162922192 cites W2473439532 @default.
• W3162922192 cites W2510725918 @default.
• W3162922192 cites W2555684706 @default.
• W3162922192 cites W2592232824 @default.
• W3162922192 cites W2593414223 @default.
• W3162922192 cites W2603777577 @default.
• W3162922192 cites W2798706078 @default.
• W3162922192 cites W2799041689 @default.
• W3162922192 cites W2809878087 @default.
• W3162922192 cites W2890680318 @default.
• W3162922192 cites W2910603373 @default.
• W3162922192 cites W2937229771 @default.
• W3162922192 cites W2954609848 @default.
• W3162922192 cites W2954996726 @default.
• W3162922192 cites W2962059918 @default.
• W3162922192 cites W2962770929 @default.
• W3162922192 cites W2962787423 @default.
• W3162922192 cites W2962793481 @default.
• W3162922192 cites W2962974533 @default.
• W3162922192 cites W2963073614 @default.
• W3162922192 cites W2963116854 @default.
• W3162922192 cites W2963349562 @default.
• W3162922192 cites W2963350032 @default.
• W3162922192 cites W2963391470 @default.
• W3162922192 cites W2963626105 @default.
• W3162922192 cites W2963767194 @default.
• W3162922192 cites W2963890275 @default.
• W3162922192 cites W2963920537 @default.
• W3162922192 cites W2964030452 @default.
• W3162922192 cites W297909767 @default.
• W3162922192 cites W2982232682 @default.
• W3162922192 cites W2984306354 @default.
• W3162922192 cites W2989855043 @default.
• W3162922192 cites W3023702633 @default.
• W3162922192 cites W3034217208 @default.
• W3162922192 cites W3034700241 @default.
• W3162922192 cites W3037771725 @default.
• W3162922192 cites W3080182903 @default.
• W3162922192 cites W3098814094 @default.
• W3162922192 cites W3125817857 @default.
• W3162922192 cites W4288359825 @default.
• W3162922192 doi "https://doi.org/10.1007/s11263-021-01448-w" @default.
• W3162922192 hasPublicationYear "2021" @default.
• W3162922192 type Work @default.
• W3162922192 sameAs 3162922192 @default.
• W3162922192 citedByCount "19" @default.
• W3162922192 countsByYear W31629221922021 @default.
• W3162922192 countsByYear W31629221922022 @default.
• W3162922192 countsByYear W31629221922023 @default.
• W3162922192 crossrefType "journal-article" @default.
• W3162922192 hasAuthorship W3162922192A5016033078 @default.
• W3162922192 hasAuthorship W3162922192A5017437981 @default.
• W3162922192 hasAuthorship W3162922192A5046779014 @default.
• W3162922192 hasAuthorship W3162922192A5050734738 @default.
• W3162922192 hasAuthorship W3162922192A5074117139 @default.
• W3162922192 hasBestOaLocation W31629221921 @default.
• W3162922192 hasConcept C119857082 @default.
• W3162922192 hasConcept C124101348 @default.
• W3162922192 hasConcept C134306372 @default.
• W3162922192 hasConcept C144024400 @default.
• W3162922192 hasConcept C153180895 @default.
• W3162922192 hasConcept C154945302 @default.
• W3162922192 hasConcept C177264268 @default.
• W3162922192 hasConcept C17744445 @default.
• W3162922192 hasConcept C199360897 @default.
• W3162922192 hasConcept C199539241 @default.
• W3162922192 hasConcept C202444582 @default.
• W3162922192 hasConcept C2776359362 @default.
• W3162922192 hasConcept C2779304628 @default.
• W3162922192 hasConcept C33923547 @default.
• W3162922192 hasConcept C36289849 @default.
• W3162922192 hasConcept C41008148 @default.
• W3162922192 hasConcept C42747912 @default.
• W3162922192 hasConcept C84392682 @default.
• W3162922192 hasConcept C94625758 @default.
• W3162922192 hasConceptScore W3162922192C119857082 @default.
• W3162922192 hasConceptScore W3162922192C124101348 @default.
• W3162922192 hasConceptScore W3162922192C134306372 @default.
• W3162922192 hasConceptScore W3162922192C144024400 @default.

• next