IEEE

Android Malware Detection Using Deep Learning

Abstract: The users got more benefits due to the rapid growth of android devices, it also increased the risk of malware attacks. Traditional methods fall short in identifying new threats, as the ...
IEEE

Unveiling Illusionary Robust Features: A Novel Approach for Adversarial Defenses in Deep Neural Networks

Abstract: Deep Neural Networks (DNNs) are vulnerable to visually imperceptible perturbations, known as Adversarial Examples (AEs). The leading hypothesis attributes this susceptibility to “non-robust ...
GitHub

autoencoder-anomaly-detection

This repo contains all my Deep Learning semester work, including implementations of FNNs, CNNs, autoencoders, CBOW, and transfer learning. I explored TensorFlow, Keras, PyTorch, and Theano while ...
earth

Why fish in the deep-sea evolve such strange shapes

The deep sea hides life in strange forms. Far below the light, pressure crushes, and temperatures drop near freezing. Yet fish thrive there, shaped by forces we barely understand. A new study reveals ...
Bleeping Computer

Decades-old ‘Finger’ protocol abused in ClickFix malware attacks

The decades-old "finger" command is making a comeback,, with threat actors using the protocol to retrieve remote commands to execute on Windows devices. In the past, people used the finger command to ...
GitHub

A New Hybrid Model for Improving Outlier Detection Using Combined Autoencoder and Variational Autoencoder (AVE)

This repository contains the source code, scripts, and supplementary materials for the paper: "A New Hybrid Model for Improving Outlier Detection Using Combined Autoencoder and Variational Autoencoder ...