FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4308641598> ?p ?o ?g. }

• W4308641598 abstract "Machine Learning (ML) software can lead to unfair and unethical decisions, making software fairness bugs an increasingly significant concern for software engineers. However, addressing fairness bugs often comes at the cost of introducing more ML performance (e.g., accuracy) bugs. In this paper, we propose MAAT, a novel ensemble approach to improving fairness-performance trade-off for ML software. Conventional ensemble methods combine different models with identical learning objectives. MAAT, instead, combines models optimized for different objectives: fairness and ML performance. We conduct an extensive evaluation of MAAT with 5 state-of-the-art methods, 9 software decision tasks, and 15 fairness-performance measurements. The results show that MAAT significantly outperforms the state-of-the-art. In particular, MAAT beats the trade-off baseline constructed by a recent benchmarking tool in 92.2% of the overall cases evaluated, 12.2 percentage points more than the best technique currently available. Moreover, the superiority of MAAT over the state-of-the-art holds on all the tasks and measurements that we study. We have made publicly available the code and data of this work to allow for future replication and extension." @default.
• W4308641598 created "2022-11-13" @default.
• W4308641598 creator A5000019783 @default.
• W4308641598 creator A5012165852 @default.
• W4308641598 creator A5031457464 @default.
• W4308641598 creator A5088708850 @default.
• W4308641598 date "2022-11-07" @default.
• W4308641598 modified "2023-09-28" @default.
• W4308641598 title "MAAT: a novel ensemble approach to addressing fairness and performance bugs for machine learning software" @default.
• W4308641598 cites W1961345416 @default.
• W4308641598 cites W1972978214 @default.
• W4308641598 cites W1979769549 @default.
• W4308641598 cites W2026019770 @default.
• W4308641598 cites W2115629999 @default.
• W4308641598 cites W2116984840 @default.
• W4308641598 cites W2150997454 @default.
• W4308641598 cites W2154602077 @default.
• W4308641598 cites W2584805976 @default.
• W4308641598 cites W2599025709 @default.
• W4308641598 cites W2730550703 @default.
• W4308641598 cites W2809701591 @default.
• W4308641598 cites W2898851569 @default.
• W4308641598 cites W2910750801 @default.
• W4308641598 cites W2913771223 @default.
• W4308641598 cites W2963116854 @default.
• W4308641598 cites W2967682612 @default.
• W4308641598 cites W2974817986 @default.
• W4308641598 cites W2982500526 @default.
• W4308641598 cites W3002398329 @default.
• W4308641598 cites W3005597817 @default.
• W4308641598 cites W3005784127 @default.
• W4308641598 cites W3008843783 @default.
• W4308641598 cites W3009290003 @default.
• W4308641598 cites W3014437350 @default.
• W4308641598 cites W3032152562 @default.
• W4308641598 cites W3089161682 @default.
• W4308641598 cites W3090943961 @default.
• W4308641598 cites W3091407209 @default.
• W4308641598 cites W3092143968 @default.
• W4308641598 cites W3093309355 @default.
• W4308641598 cites W3098260236 @default.
• W4308641598 cites W3099095494 @default.
• W4308641598 cites W3103264664 @default.
• W4308641598 cites W3103741452 @default.
• W4308641598 cites W3122215046 @default.
• W4308641598 cites W3123477784 @default.
• W4308641598 cites W3124540000 @default.
• W4308641598 cites W3128630643 @default.
• W4308641598 cites W3130354627 @default.
• W4308641598 cites W3152436735 @default.
• W4308641598 cites W3175995826 @default.
• W4308641598 cites W3176008422 @default.
• W4308641598 cites W3179976352 @default.
• W4308641598 cites W3193448347 @default.
• W4308641598 cites W3194157648 @default.
• W4308641598 cites W3194588521 @default.
• W4308641598 cites W3216660278 @default.
• W4308641598 cites W4206787394 @default.
• W4308641598 doi "https://doi.org/10.1145/3540250.3549093" @default.
• W4308641598 hasPublicationYear "2022" @default.
• W4308641598 type Work @default.
• W4308641598 citedByCount "12" @default.
• W4308641598 countsByYear W43086415982023 @default.
• W4308641598 crossrefType "proceedings-article" @default.
• W4308641598 hasAuthorship W4308641598A5000019783 @default.
• W4308641598 hasAuthorship W4308641598A5012165852 @default.
• W4308641598 hasAuthorship W4308641598A5031457464 @default.
• W4308641598 hasAuthorship W4308641598A5088708850 @default.
• W4308641598 hasBestOaLocation W43086415982 @default.
• W4308641598 hasConcept C1009929 @default.
• W4308641598 hasConcept C105795698 @default.
• W4308641598 hasConcept C111368507 @default.
• W4308641598 hasConcept C111919701 @default.
• W4308641598 hasConcept C115903868 @default.
• W4308641598 hasConcept C119857082 @default.
• W4308641598 hasConcept C12590798 @default.
• W4308641598 hasConcept C12725497 @default.
• W4308641598 hasConcept C127313418 @default.
• W4308641598 hasConcept C144133560 @default.
• W4308641598 hasConcept C154945302 @default.
• W4308641598 hasConcept C162853370 @default.
• W4308641598 hasConcept C177264268 @default.
• W4308641598 hasConcept C199360897 @default.
• W4308641598 hasConcept C2776760102 @default.
• W4308641598 hasConcept C2777904410 @default.
• W4308641598 hasConcept C33923547 @default.
• W4308641598 hasConcept C41008148 @default.
• W4308641598 hasConcept C45942800 @default.
• W4308641598 hasConcept C86251818 @default.
• W4308641598 hasConceptScore W4308641598C1009929 @default.
• W4308641598 hasConceptScore W4308641598C105795698 @default.
• W4308641598 hasConceptScore W4308641598C111368507 @default.
• W4308641598 hasConceptScore W4308641598C111919701 @default.
• W4308641598 hasConceptScore W4308641598C115903868 @default.
• W4308641598 hasConceptScore W4308641598C119857082 @default.
• W4308641598 hasConceptScore W4308641598C12590798 @default.
• W4308641598 hasConceptScore W4308641598C12725497 @default.
• W4308641598 hasConceptScore W4308641598C127313418 @default.
• W4308641598 hasConceptScore W4308641598C144133560 @default.
• W4308641598 hasConceptScore W4308641598C154945302 @default.

• next