%0 Generic %1 sundararajan2017axiomatic %A Sundararajan, Mukund %A Taly, Ankur %A Yan, Qiqi %D 2017 %K machine-learning %T Axiomatic Attribution for Deep Networks %U http://arxiv.org/abs/1703.01365 %X We study the problem of attributing the prediction of a deep network to its input features, a problem previously studied by several other works. We identify two fundamental axioms---Sensitivity and Implementation Invariance that attribution methods ought to satisfy. We show that they are not satisfied by most known attribution methods, which we consider to be a fundamental weakness of those methods. We use the axioms to guide the design of a new attribution method called Integrated Gradients. Our method requires no modification to the original network and is extremely simple to implement; it just needs a few calls to the standard gradient operator. We apply this method to a couple of image models, a couple of text models and a chemistry model, demonstrating its ability to debug networks, to extract rules from a network, and to enable users to engage with models better.

@misc{sundararajan2017axiomatic, abstract = {We study the problem of attributing the prediction of a deep network to its input features, a problem previously studied by several other works. We identify two fundamental axioms---Sensitivity and Implementation Invariance that attribution methods ought to satisfy. We show that they are not satisfied by most known attribution methods, which we consider to be a fundamental weakness of those methods. We use the axioms to guide the design of a new attribution method called Integrated Gradients. Our method requires no modification to the original network and is extremely simple to implement; it just needs a few calls to the standard gradient operator. We apply this method to a couple of image models, a couple of text models and a chemistry model, demonstrating its ability to debug networks, to extract rules from a network, and to enable users to engage with models better.}, added-at = {2022-10-24T19:25:56.000+0200}, author = {Sundararajan, Mukund and Taly, Ankur and Yan, Qiqi}, biburl = {https://www.bibsonomy.org/bibtex/2068847d63e30d9f49ef0ecb021f5cabb/jpbarrettel}, description = {Axiomatic Attribution for Deep Networks}, interhash = {aff51bf8900a6904e8f7357668dc369c}, intrahash = {068847d63e30d9f49ef0ecb021f5cabb}, keywords = {machine-learning}, note = {cite arxiv:1703.01365}, timestamp = {2022-10-24T19:25:56.000+0200}, title = {Axiomatic Attribution for Deep Networks}, url = {http://arxiv.org/abs/1703.01365}, year = 2017 }