Models on the Move: Towards Feasible Embedded AI for Intrusion Detection on Vehicular CAN Bus
Offered By: USENIX via YouTube
Course Description
Overview
Explore a conference talk that delves into the development of an efficient and effective intrusion detection system (IDS) for vehicular Controller Area Network (CAN) bus security. Learn about the MULSAM model, which employs multi-dimensional long short-term memory with a self-attention mechanism to address the challenge of small-batch attacks. Discover how this innovative approach achieves superior training stability and detection accuracy for identifying small-batch injection attacks on vehicular CAN bus systems. Gain insights into the hardware implementation of MULSAM as an embedded unit in vehicles, including its FPGA-based accelerator that offers improved energy efficiency and low latency while maintaining high detection accuracy. Understand the potential of this cyber-physical system security solution in advancing embedded AI for intrusion detection in modern vehicles.
Syllabus
USENIX ATC '24 - Models on the Move: Towards Feasible Embedded AI for Intrusion Detection on...
Taught by
USENIX
Related Courses
Reinforcement Learning for Trading StrategiesNew York Institute of Finance via Coursera Natural Language Processing with Sequence Models
DeepLearning.AI via Coursera Fake News Detection with Machine Learning
Coursera Project Network via Coursera English/French Translator: Long Short Term Memory Networks
Coursera Project Network via Coursera Text Classification Using Word2Vec and LSTM on Keras
Coursera Project Network via Coursera