Matches in SemOpenAlex for { <https://semopenalex.org/work/W4295267831> ?p ?o ?g. }
Showing items 1 to 59 of
59
with 100 items per page.
- W4295267831 abstract "Shapley values underlie one of the most popular model-agnostic methods within explainable artificial intelligence. These values are designed to attribute the difference between a model's prediction and an average baseline to the different features used as input to the model. Being based on solid game-theoretic principles, Shapley values uniquely satisfy several desirable properties, which is why they are increasingly used to explain the predictions of possibly complex and highly non-linear machine learning models. Shapley values are well calibrated to a user's intuition when features are independent, but may lead to undesirable, counterintuitive explanations when the independence assumption is violated. In this paper, we propose a novel framework for computing Shapley values that generalizes recent work that aims to circumvent the independence assumption. By employing Pearl's do-calculus, we show how these 'causal' Shapley values can be derived for general causal graphs without sacrificing any of their desirable properties. Moreover, causal Shapley values enable us to separate the contribution of direct and indirect effects. We provide a practical implementation for computing causal Shapley values based on causal chain graphs when only partial information is available and illustrate their utility on a real-world example." @default.
- W4295267831 created "2022-09-12" @default.
- W4295267831 creator A5005492854 @default.
- W4295267831 creator A5015444349 @default.
- W4295267831 creator A5052460567 @default.
- W4295267831 creator A5068651384 @default.
- W4295267831 date "2020-11-03" @default.
- W4295267831 modified "2023-10-14" @default.
- W4295267831 title "Causal Shapley Values: Exploiting Causal Knowledge to Explain Individual Predictions of Complex Models" @default.
- W4295267831 doi "https://doi.org/10.48550/arxiv.2011.01625" @default.
- W4295267831 hasPublicationYear "2020" @default.
- W4295267831 type Work @default.
- W4295267831 citedByCount "0" @default.
- W4295267831 crossrefType "posted-content" @default.
- W4295267831 hasAuthorship W4295267831A5005492854 @default.
- W4295267831 hasAuthorship W4295267831A5015444349 @default.
- W4295267831 hasAuthorship W4295267831A5052460567 @default.
- W4295267831 hasAuthorship W4295267831A5068651384 @default.
- W4295267831 hasBestOaLocation W42952678311 @default.
- W4295267831 hasConcept C101097943 @default.
- W4295267831 hasConcept C105795698 @default.
- W4295267831 hasConcept C111472728 @default.
- W4295267831 hasConcept C11671645 @default.
- W4295267831 hasConcept C132010649 @default.
- W4295267831 hasConcept C138885662 @default.
- W4295267831 hasConcept C144237770 @default.
- W4295267831 hasConcept C177142836 @default.
- W4295267831 hasConcept C199022921 @default.
- W4295267831 hasConcept C33923547 @default.
- W4295267831 hasConcept C35651441 @default.
- W4295267831 hasConcept C41008148 @default.
- W4295267831 hasConceptScore W4295267831C101097943 @default.
- W4295267831 hasConceptScore W4295267831C105795698 @default.
- W4295267831 hasConceptScore W4295267831C111472728 @default.
- W4295267831 hasConceptScore W4295267831C11671645 @default.
- W4295267831 hasConceptScore W4295267831C132010649 @default.
- W4295267831 hasConceptScore W4295267831C138885662 @default.
- W4295267831 hasConceptScore W4295267831C144237770 @default.
- W4295267831 hasConceptScore W4295267831C177142836 @default.
- W4295267831 hasConceptScore W4295267831C199022921 @default.
- W4295267831 hasConceptScore W4295267831C33923547 @default.
- W4295267831 hasConceptScore W4295267831C35651441 @default.
- W4295267831 hasConceptScore W4295267831C41008148 @default.
- W4295267831 hasLocation W42952678311 @default.
- W4295267831 hasOpenAccess W4295267831 @default.
- W4295267831 hasPrimaryLocation W42952678311 @default.
- W4295267831 hasRelatedWork W1562844845 @default.
- W4295267831 hasRelatedWork W1593369943 @default.
- W4295267831 hasRelatedWork W1973141406 @default.
- W4295267831 hasRelatedWork W2042580993 @default.
- W4295267831 hasRelatedWork W2043333589 @default.
- W4295267831 hasRelatedWork W2092695596 @default.
- W4295267831 hasRelatedWork W2885187933 @default.
- W4295267831 hasRelatedWork W3165895314 @default.
- W4295267831 hasRelatedWork W4281561852 @default.
- W4295267831 hasRelatedWork W2143549598 @default.
- W4295267831 isParatext "false" @default.
- W4295267831 isRetracted "false" @default.
- W4295267831 workType "article" @default.