Oleg Zabluda's blog
Wednesday, March 14, 2018
 
Axiomatic Attribution for Deep Networks (2017) Mukund Sundararajan, Ankur Taly, Qiqi Yan
Axiomatic Attribution for Deep Networks (2017) Mukund Sundararajan, Ankur Taly, Qiqi Yan
"""
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.
[...]
Theorem 1. Integrated gradients is the unique path method that is symmetry-preserving.
"""
https://arxiv.org/abs/1703.01365
https://arxiv.org/abs/1703.01365

Labels:



Powered by Blogger