Matches in SemOpenAlex for { <https://semopenalex.org/work/W2983048110> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W2983048110 abstract "Fair classification has become an important topic in machine learning research. While most bias mitigation strategies focus on neural networks, we noticed a lack of work on fair classifiers based on decision trees even though they have proven very efficient. In an up-to-date comparison of state-of-the-art classification algorithms in tabular data, tree boosting outperforms deep learning. For this reason, we have developed a novel approach of adversarial gradient tree boosting. The objective of the algorithm is to predict the output $Y$ with gradient tree boosting while minimizing the ability of an adversarial neural network to predict the sensitive attribute $S$. The approach incorporates at each iteration the gradient of the neural network directly in the gradient tree boosting. We empirically assess our approach on 4 popular data sets and compare against state-of-the-art algorithms. The results show that our algorithm achieves a higher accuracy while obtaining the same level of fairness, as measured using a set of different common fairness definitions." @default.
- W2983048110 created "2019-11-22" @default.
- W2983048110 creator A5008031795 @default.
- W2983048110 creator A5031635996 @default.
- W2983048110 creator A5039251226 @default.
- W2983048110 creator A5065404842 @default.
- W2983048110 date "2019-11-13" @default.
- W2983048110 modified "2023-09-25" @default.
- W2983048110 title "Fair Adversarial Gradient Tree Boosting" @default.
- W2983048110 cites W1961345416 @default.
- W2983048110 cites W2014352947 @default.
- W2983048110 cites W2085988980 @default.
- W2983048110 cites W2099471712 @default.
- W2983048110 cites W2100960835 @default.
- W2983048110 cites W2130486630 @default.
- W2983048110 cites W2530395818 @default.
- W2983048110 cites W2540757487 @default.
- W2983048110 cites W2546563948 @default.
- W2983048110 cites W2604504584 @default.
- W2983048110 cites W2732560823 @default.
- W2983048110 cites W2810290439 @default.
- W2983048110 cites W2963100392 @default.
- W2983048110 cites W2963116854 @default.
- W2983048110 cites W3023309920 @default.
- W2983048110 cites W3120740533 @default.
- W2983048110 hasPublicationYear "2019" @default.
- W2983048110 type Work @default.
- W2983048110 sameAs 2983048110 @default.
- W2983048110 citedByCount "2" @default.
- W2983048110 countsByYear W29830481102020 @default.
- W2983048110 countsByYear W29830481102021 @default.
- W2983048110 crossrefType "posted-content" @default.
- W2983048110 hasAuthorship W2983048110A5008031795 @default.
- W2983048110 hasAuthorship W2983048110A5031635996 @default.
- W2983048110 hasAuthorship W2983048110A5039251226 @default.
- W2983048110 hasAuthorship W2983048110A5065404842 @default.
- W2983048110 hasConcept C10229987 @default.
- W2983048110 hasConcept C113174947 @default.
- W2983048110 hasConcept C119857082 @default.
- W2983048110 hasConcept C120136583 @default.
- W2983048110 hasConcept C124101348 @default.
- W2983048110 hasConcept C134306372 @default.
- W2983048110 hasConcept C154945302 @default.
- W2983048110 hasConcept C169258074 @default.
- W2983048110 hasConcept C33923547 @default.
- W2983048110 hasConcept C37736160 @default.
- W2983048110 hasConcept C41008148 @default.
- W2983048110 hasConcept C46686674 @default.
- W2983048110 hasConcept C50644808 @default.
- W2983048110 hasConcept C5481197 @default.
- W2983048110 hasConcept C70153297 @default.
- W2983048110 hasConcept C84525736 @default.
- W2983048110 hasConceptScore W2983048110C10229987 @default.
- W2983048110 hasConceptScore W2983048110C113174947 @default.
- W2983048110 hasConceptScore W2983048110C119857082 @default.
- W2983048110 hasConceptScore W2983048110C120136583 @default.
- W2983048110 hasConceptScore W2983048110C124101348 @default.
- W2983048110 hasConceptScore W2983048110C134306372 @default.
- W2983048110 hasConceptScore W2983048110C154945302 @default.
- W2983048110 hasConceptScore W2983048110C169258074 @default.
- W2983048110 hasConceptScore W2983048110C33923547 @default.
- W2983048110 hasConceptScore W2983048110C37736160 @default.
- W2983048110 hasConceptScore W2983048110C41008148 @default.
- W2983048110 hasConceptScore W2983048110C46686674 @default.
- W2983048110 hasConceptScore W2983048110C50644808 @default.
- W2983048110 hasConceptScore W2983048110C5481197 @default.
- W2983048110 hasConceptScore W2983048110C70153297 @default.
- W2983048110 hasConceptScore W2983048110C84525736 @default.
- W2983048110 hasLocation W29830481101 @default.
- W2983048110 hasOpenAccess W2983048110 @default.
- W2983048110 hasPrimaryLocation W29830481101 @default.
- W2983048110 hasRelatedWork W2032210760 @default.
- W2983048110 hasRelatedWork W2065684080 @default.
- W2983048110 hasRelatedWork W2125048840 @default.
- W2983048110 hasRelatedWork W2137918313 @default.
- W2983048110 hasRelatedWork W2154225404 @default.
- W2983048110 hasRelatedWork W2465306788 @default.
- W2983048110 hasRelatedWork W2494495812 @default.
- W2983048110 hasRelatedWork W2515758957 @default.
- W2983048110 hasRelatedWork W2608208138 @default.
- W2983048110 hasRelatedWork W2724434755 @default.
- W2983048110 hasRelatedWork W2791407956 @default.
- W2983048110 hasRelatedWork W2908245698 @default.
- W2983048110 hasRelatedWork W2911616853 @default.
- W2983048110 hasRelatedWork W2912071732 @default.
- W2983048110 hasRelatedWork W2952714963 @default.
- W2983048110 hasRelatedWork W3004246571 @default.
- W2983048110 hasRelatedWork W3114774860 @default.
- W2983048110 hasRelatedWork W3119028626 @default.
- W2983048110 hasRelatedWork W46572615 @default.
- W2983048110 hasRelatedWork W2336541324 @default.
- W2983048110 isParatext "false" @default.
- W2983048110 isRetracted "false" @default.
- W2983048110 magId "2983048110" @default.
- W2983048110 workType "article" @default.