Clo-HDnn: A 4.66 TFLOPS/W and 3.78 TOPS/W Continual On-Device Learning Accelerator with Energy-efficient Hyperdimensional Computing via Progressive Search
Clo-HDnn is an on-device learning (ODL) accelerator designed for emerging continual learning (CL) tasks. Clo-HDnn integrates hyperdimensional computing (HDC) along with low-cost Kronecker HD Encoder and weight clustering feature extraction (WCFE) to optimize accuracy and efficiency. Clo-HDnn adopts gradient-free CL to efficiently update and store the learned knowledge in the form of class hypervectors. Its dual-mode operation enables bypassing costly feature extraction for simpler datasets, while progressive search reduces complexity by up to 61% by encoding and comparing only partial query hypervectors. Achieving 4.66 TFLOPS/W (FE) and 3.78 TOPS/W (classifier), Clo-HDnn delivers 7.77x and 4.85x higher energy efficiency compared to SOTA ODL accelerators.
@article{arxiv.2507.17953,
title = {Clo-HDnn: A 4.66 TFLOPS/W and 3.78 TOPS/W Continual On-Device Learning Accelerator with Energy-efficient Hyperdimensional Computing via Progressive Search},
author = {Chang Eun Song and Weihong Xu and Keming Fan and Soumil Jain and Gopabandhu Hota and Haichao Yang and Leo Liu and Kerem Akarvardar and Meng-Fan Chang and Carlos H. Diaz and Gert Cauwenberghs and Tajana Rosing and Mingu Kang},
journal= {arXiv preprint arXiv:2507.17953},
year = {2025}
}
Comments
Published in 2025 Symposium on VLSI Technology and Circuits (VLSI Technology and Circuits), Kyoto, Japan, 2025