Explore a cutting-edge approach to detecting memory-resident malware using convolution neural networks in this Black Hat conference talk. Delve into the Mem2Img framework, designed to overcome the limitations of traditional antivirus software and YARA rules in identifying unknown and shellcode-based malware. Learn how this innovative technique leverages machine learning to classify malware families and detect threats without relying on easily evaded handcrafted features. Discover the potential of this method to improve invisibility and achieve persistence in the face of advanced persistent threats (APTs) and process injection techniques. Gain insights from security experts Charles Li and Aragorn Tseng as they present their research on enhancing malware detection capabilities for better cybersecurity defense.
Mem2Img - Memory-Resident Malware Detection via Convolution Neural Network
Overview
Syllabus
Mem2Img: Memory-Resident Malware Detection via Convolution Neural Network
Taught by
Black Hat
Related Courses
-
Static Detection of Novel Malware Using Transfer Learning with Deep Neural Networks
-
Machine Learning for Enhanced Malware Detection and Classification
-
Malware Classification With Machine Learning Enhanced by Windows Kernel Emulation
-
Deep in the Dark - Deep Learning-based Malware Traffic Detection without Expert Knowledge
-
New Memory Forensics Techniques to Defeat Device Monitoring Malware
-
Deep Learning for Realtime Malware Detection
