ملف الباحث
Sumohana S. Channappayya
ورقتان في مجموعة PaperMetrix
المنشورات
أوراق هذا المؤلف
-
Perceptually-Inspired Local Source Normalization for Adversarial Robustness
2023
The importance of the robustness of deep neural networks (DNNs) cannot be overemphasised due to their widespread use in safety-critical applications. However, the fragility of DNNs to imperceptible input perturbations is well-known in the literature. …
-
Minimizing Energy Costs in Deep Learning Model Training: The Gaussian Sampling Approach
2024 · arXiv (Cornell University)
Computing the loss gradient via backpropagation consumes considerable energy during deep learning (DL) model training. In this paper, we propose a novel approach to efficiently compute DL models' gradients to mitigate the substantial energy overhead …