| Perceptual Adversarial Robustness: Defense Against Unseen Threat Models C Laidlaw, S Singla, S Feizi International Conference on Learning Representations (ICLR), 2021, 2020 | 73 | 2020 |
| Understanding Impacts of High-Order Loss Approximations and Features in Deep Learning Interpretation S Singla, E Wallace, S Feng, S Feizi International Conference on Machine Learning (ICML), 2019, 0 | 32* | |
| Second-order provable defenses against adversarial attacks S Singla, S Feizi International Conference on Machine Learning (ICML), 2020, 0 | 31* | |
| Fairness Through Robustness: Investigating Robustness Disparity in Deep Learning V Nanda, S Dooley, S Singla, S Feizi, JP Dickerson Conference on Fairness, Accountability, and Transparency (FAccT), 2021, 466-477, 2021 | 27 | 2021 |
| Certifiably robust interpretation in deep learning A Levine, S Singla, S Feizi arXiv preprint arXiv:1905.12105, 2019 | 26 | 2019 |
| Understanding Failures of Deep Networks via Robust Feature Extraction S Singla, B Nushi, S Shah, E Kamar, E Horvitz Computer Vision and Pattern Recognition (CVPR), 2021, 12853-12862, 0 | 16* | |
| Skew Orthogonal Convolutions S Singla, S Feizi International Conference on Machine Learning (ICML), 2021, 0 | 14* | |
| Fantastic Four: Differentiable Bounds on Singular Values of Convolution Layers S Singla, S Feizi arXiv preprint arXiv:1911.10258, 2019 | 12 | 2019 |
| Low Curvature Activations Reduce Overfitting in Adversarial Training V Singla, S Singla, D Jacobs, S Feizi International Conference on Computer Vision (ICCV), 2021, 2021 | 11 | 2021 |
| Robustness certificates against adversarial examples for relu networks S Singla, S Feizi arXiv preprint arXiv:1902.01235, 2019 | 11 | 2019 |
| Fantastic four: Differentiable and efficient bounds on singular values of convolution layers S Singla, S Feizi International Conference on Learning Representations (ICLR), 2021, 2020 | 6 | 2020 |
| Salient ImageNet: How to discover spurious features in deep learning? SF Sahil Singla International Conference on Learning Representations, 2022, 0 | | |
Soheil FeiziUniversity of Maryland, College ParkVerified email at cs.umd.edu
