Improving Attribution Methods by Learning Submodular Functions
arXiv:2104.09073
Abstract
This work explores the novel idea of learning a submodular scoring function to improve the specificity/selectivity of existing feature attribution methods. Submodular scores are natural for attribution as they are known to accurately model the principle of diminishing returns. A new formulation for learning a deep submodular set function that is consistent with the real-valued attribution maps obtained by existing attribution methods is proposed. The final attribution value of a feature is then defined as the marginal gain in the induced submodular score of the feature in the context of other highly attributed features, thus decreasing the attribution of redundant yet discriminatory features. Experiments on multiple datasets illustrate that the proposed attribution method achieves higher specificity along with good discriminative power. The implementation of our method is publicly available at https://github.com/Piyushi-0/SEA-NN.
References in corpus (13)
- Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networks
- Axiomatic Attribution for Deep Networks
- Learning Important Features Through Propagating Activation Differences
- Explaining NonLinear Classification Decisions with Deep Taylor Decomposition
- SmoothGrad: removing noise by adding noise
- Visualizing and Understanding Convolutional Networks
- Towards Robust Interpretability with Self-Explaining Neural Networks
- A Benchmark for Interpretability Methods in Deep Neural Networks
- Grad-CAM: Why did you say that?
- RISE: Randomized Input Sampling for Explanation of Black-box Models
- A Human-Grounded Evaluation Benchmark for Local Explanations of Machine Learning
- Understanding Impacts of High-Order Loss Approximations and Features in Deep Learning Interpretation
- A simple defense against adversarial attacks on heatmap explanations