TowardsRigorousInterpretations:aFormalisationofFeatureAttributionDariusAfchar12RomainHennequin1VincentGuigue2Abstractcontrollinganddisentanglingeveryaspectofamodelpre-diction,rangingfromhowthedatai...
InterpretationsareUseful:PenalizingExplanationstoAlignNeuralNetworkswithPriorKnowledgeLauraRieger1ChandanSingh2W.JamesMurdoch3BinYu23Abstractimproveamodel.ForanexplanationofadeeplearningmodeltoPred...