D2H-AD — hyperdimensional computing for anomaly detection
arXiv paper proposes D2H-AD, an anomaly detection framework based on hyperdimensional computing (HDC), a brain-inspired method that uses high-dimensional vectors to represent data.
Designed to work with limited labeled data and minimal compute — targets edge devices, IoT, healthcare, and cybersecurity use cases where conventional ML and deep learning break down.
Combines distance-based similarity and density-aware encoding in a single unified model, moving beyond prior HDC-only anomaly detection work.