Výsledky vyhledávání - "Rahman, Jamshaid Ul"

Report

DiffGrad for Physics-Informed Neural Networks

Autor: Rahman, Jamshaid Ul, Nimra

Physics-Informed Neural Networks (PINNs) are regarded as state-of-the-art tools for addressing highly nonlinear problems based on partial differential equations. Despite their broad range of applications, PINNs encounter several performance challenge

Externí odkaz: http://arxiv.org/abs/2409.03239

Zobrazit plný text záznamu

Report

SwishReLU: A Unified Approach to Activation Functions for Enhanced Deep Neural Networks Performance

Autor: Rahman, Jamshaid Ul, Zulfiqar, Rubiqa, Khan, Asad, Nimra

ReLU, a commonly used activation function in deep neural networks, is prone to the issue of "Dying ReLU". Several enhanced versions, such as ELU, SeLU, and Swish, have been introduced and are considered to be less commonly utilized. However, replacin

Externí odkaz: http://arxiv.org/abs/2407.08232

Zobrazit plný text záznamu

Report

ASU-CNN: An Efficient Deep Architecture for Image Classification and Feature Visualizations

Autor: Rahman, Jamshaid Ul, Makhdoom, Faiza, Lu, Dianchen

Activation functions play a decisive role in determining the capacity of Deep Neural Networks as they enable neural networks to capture inherent nonlinearities present in data fed to them. The prior research on activation functions primarily focused

Externí odkaz: http://arxiv.org/abs/2305.19146

Zobrazit plný text záznamu

Report

Amplifying Sine Unit: An Oscillatory Activation Function for Deep Neural Networks to Recover Nonlinear Oscillations Efficiently

Autor: Rahman, Jamshaid Ul, Makhdoom, Faiza, Lu, Dianchen

Many industrial and real life problems exhibit highly nonlinear periodic behaviors and the conventional methods may fall short of finding their analytical or closed form solutions. Such problems demand some cutting edge computational tools with incre

Externí odkaz: http://arxiv.org/abs/2304.09759

Zobrazit plný text záznamu