OptimalCounterfactualExplanationsinTreeEnsemblesAxelParmentier1ThibautVidal23AbstracttriggeredextensiveresearchonalgorithmicrecourseandcounterfactualExplanations(see,e.g.,Wachteretal.,2018;Counterf...
GenerativeCausalExplanationsforGraphNeuralNetworksWanyuLin12HaoLan2BaochunLi2AbstractDespitetheirpracticalsuccess,mostGNNsaredeployedasblackboxes,lackingexplicitdeclarativeknowledgerepre-Thispaperp...
ExplanationsforMonotonicClassifiersJoaoMarques-Silva1ThomasGerspacher1MartinCooper1AlexeyIgnatiev2NinaNarodytska3Abstract2018),activelearning(Barile&Feelders,2012)and,morerecently,infairness(Wang&G...
RobustandStableBlackBoxExplanationsHimabinduLakkaraju1NinoArsov2OsbertBastani3Abstractanddecidewhenandhowtoemploythem.However,theproprietarynatureandincreasingcomplexityofmachineAsmachinelearningbl...
ProblemswithShapley-value-basedExplanationsasfeatureimportancemeasuresI.ElizabethKumar1SureshVenkatasubramanian1CarlosScheidegger2SorelleA.Friedler3Abstractbythemodel,andthegameisthepredictionofthe...
InterpretationsareUseful:PenalizingExplanationstoAlignNeuralNetworkswithPriorKnowledgeLauraRieger1ChandanSingh2W.JamesMurdoch3BinYu23Abstractimproveamodel.ForanexplanationofadeeplearningmodeltoPred...
FairwashingExplanationswithOff-ManifoldDetergentChristopherJ.Anders1PlamenPasliev1Ann-KathrinDombrowski1Klaus-RobertMu¨ller123PanKessel1Abstractquestionablefeatures.Itcanfurthermoreprovidefeedback...
ConciseExplanationsofNeuralNetworksusingAdversarialTrainingPrasadChalasani1JiefengChen2AmritaRoyChowdhury2SomeshJha12XiWu3Abstract1.IntroductionWeshownewconnectionsbetweenadversarialDespitetherecen...
CounterfactualVisualExplanationsYashGoyal1ZiyanWu2JanErnst2DhruvBatra1DeviParikh1StefanLee1AbstractFigure1.Ourapproachgeneratescounterfactualvisualexplana-tionsforaqueryimageI(left)–explainingwhyt...
