Investigating Saturation Effects in Integrated Gradients
Abstract
Integrated Gradients has become a popular method for post-hoc model interpretability. De-spite its popularity, the composition and relative impact of different regions of the integral path are not well understood. We explore these effects and find that gradients in saturated regions of this path, where model output changes minimally, contribute disproportionately to the computed attribution. We propose a variant of IntegratedGradients which primarily captures gradients in unsaturated regions and evaluate this method on ImageNet classification networks. We find that this attribution technique shows higher model faithfulness and lower sensitivity to noise com-pared with standard Integrated Gradients. A note-book illustrating our computations and results is available at https://github.com/vivekmig/captum-1/tree/ExpandedIG.
Cite
@article{arxiv.2010.12697,
title = {Investigating Saturation Effects in Integrated Gradients},
author = {Vivek Miglani and Narine Kokhlikyan and Bilal Alsallakh and Miguel Martin and Orion Reblitz-Richardson},
journal= {arXiv preprint arXiv:2010.12697},
year = {2020}
}
Comments
Presented at ICML Workshop on Human Interpretability in Machine Learning (WHI 2020)