Research map / Adversarial robustness and security / Certified and empirical robustness
On the Connection Between Adversarial Robustness and Saliency Map Interpretability
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Increase local Lipschitz regularization strength | MNIST classifiers trained with double backpropagation · MNIST validation set | median alignment | Helped |
| Increase local Lipschitz regularization strength | ResNet50 classifiers fine-tuned with double backpropagation · ImageNet validation set | median alignment | Helped |
| Increase robustness through local Lipschitz regularization | MNIST classifiers trained with double backpropagation · MNIST validation set | median alignment | Mixed |
| Compare pointwise alignment with robustness as model robustness increases | MNIST classifiers trained with double backpropagation · MNIST validation set | pointwise correlation between alignment and linearized robustness | Helped |
| Compare pointwise alignment with robustness as model robustness increases | ResNet50 classifiers fine-tuned with double backpropagation · ImageNet validation set | pointwise correlation between alignment and linearized robustness | No clear effect |
| Use linearized robustness to estimate adversarial robustness | Neural network classifiers · ImageNet and MNIST validation sets | correlation with adversarial robustness | Helped |
Closest papers by meaning
- On the explainable properties of 1-Lipschitz Neural Networks: An Optimal Transport Perspective (NeurIPS 2023 · similarity 0.89)
- Proper Network Interpretability Helps Adversarial Robustness in Classification (ICML 2020 · similarity 0.88)
- First-Order Adversarial Vulnerability of Neural Networks and Input Dimension (ICML 2019 · similarity 0.87)
- Structured Gradient-based Interpretations via Norm-Regularized Adversarial Training (CVPR 2024 · similarity 0.87)
- Building Reliable Explanations of Unreliable Neural Networks: Locally Smoothing Perspective of Model Interpretation (CVPR 2021 · similarity 0.87)
- Great Minds Think Alike: The Universal Convergence Trend of Input Salience (NeurIPS 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.