Search Results for author:
Found 4 papers, 3 papers with code
STR-Cert: Robustness Certification for Deep Text Recognition on Deep Learning Pipelines and Vision Transformers
Robustness certification, which aims to formally certify the predictions of neural networks against adversarial inputs, has become an integral part of important tool for safety-critical applications.
Effective Structural Encodings via Local Curvature Profiles
We further show that combining local structural encodings, such as LCP, with global positional encodings improves downstream performance, suggesting that they capture complementary geometric information.
Mitigating Over-Smoothing and Over-Squashing using Augmentations of Forman-Ricci Curvature
Several rewiring approaches utilizing graph characteristics, such as curvature or the spectrum of the graph Laplacian, have been proposed.
Understanding and Mitigating Extrapolation Failures in Physics-Informed Neural Networks
Physics-informed Neural Networks (PINNs) have recently gained popularity due to their effective approximation of partial differential equations (PDEs) using deep neural networks (DNNs).
