Proceedings

On Smoother Attributions using Neural Stochastic Differential Equations

Sumit Jha, Rickard Ewetz, Alvaro Velasquez, Susmit Jha

Download BibTeX Cite

Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence. Main Track. Pages 522-528.

https://doi.org/10.24963/ijcai.2021/73

Keywords

  • AI Ethics, Trust, Fairness: Explainability
  • Multidisciplinary Topics and Applications: Validation and Verification

Abstract

Several methods have recently been developed for computing attributions of a neural network's prediction over the input features. However, these existing approaches for computing attributions are noisy and not robust to small perturbations of the input. This paper uses the recently identified connection between dynamical systems and residual neural networks to show that the attributions computed over neural stochastic differential equations (SDEs) are less noisy, visually sharper, and quantitatively more robust. Using dynamical systems theory, we theoretically analyze the robustness of these attributions. We also experimentally demonstrate the efficacy of our approach in providing smoother, visually sharper and quantitatively robust attributions by computing attributions for ImageNet images using ResNet-50, WideResNet-101 models and ResNeXt-101 models.

Sumit Jha, Rickard Ewetz, Alvaro Velasquez, and Susmit Jha. 2021. On Smoother Attributions using Neural Stochastic Differential Equations. In Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21. IJCAI, 522–528. https://doi.org/10.24963/ijcai.2021/73
Welcome Future Conferences AI Journal Division IJCAI Proceedings About IJCAI IJCAI Awards