FSL-HDnn: A 40 nm Few-shot On-Device Learning Accelerator with Integrated Feature Extraction and Hyperdimensional Computing

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Xu, Weihong, Song, Chang Eun, Yang, Haichao, Liu, Leo, Chang, Meng-Fan, Diaz, Carlos H., Rosing, Tajana, Kang, Mingu
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911316480884736
author Xu, Weihong
Song, Chang Eun
Yang, Haichao
Liu, Leo
Chang, Meng-Fan
Diaz, Carlos H.
Rosing, Tajana
Kang, Mingu
author_facet Xu, Weihong
Song, Chang Eun
Yang, Haichao
Liu, Leo
Chang, Meng-Fan
Diaz, Carlos H.
Rosing, Tajana
Kang, Mingu
contents This paper introduces FSL-HDnn, an energy-efficient accelerator that implements the end-to-end pipeline of feature extraction and on-device few-shot learning (FSL). The accelerator addresses fundamental challenges of on-device learning (ODL) for resource-constrained edge applications through two synergistic modules: a parameter-efficient feature extractor employing weight clustering and an FSL classifier based on hyperdimensional computing (HDC). The feature extractor exploits the weight clustering mechanism to reduce computational complexity, while the HDC-based FSL classifier eliminates gradient-based back propagation operations, enabling single-pass training with substantially reduced latency. Additionally, FSL-HDnn enables low-latency ODL and inference via two proposed optimization strategies, including an early-exit mechanism with branch feature extraction and batched single-pass training that improves hardware utilization. Measurement results demonstrate that our chip fabricated in a 40 nm CMOS process delivers superior training energy efficiency of 6 mJ/image and end-to-end training throughput of 28 images/s on a 10-way 5-shot FSL task. The end-to-end training latency is also reduced by 2x to 20.9x compared to state-of-the-art ODL chips.
format Preprint
id arxiv_https___arxiv_org_abs_2512_11826
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FSL-HDnn: A 40 nm Few-shot On-Device Learning Accelerator with Integrated Feature Extraction and Hyperdimensional Computing
Xu, Weihong
Song, Chang Eun
Yang, Haichao
Liu, Leo
Chang, Meng-Fan
Diaz, Carlos H.
Rosing, Tajana
Kang, Mingu
Hardware Architecture
Image and Video Processing
This paper introduces FSL-HDnn, an energy-efficient accelerator that implements the end-to-end pipeline of feature extraction and on-device few-shot learning (FSL). The accelerator addresses fundamental challenges of on-device learning (ODL) for resource-constrained edge applications through two synergistic modules: a parameter-efficient feature extractor employing weight clustering and an FSL classifier based on hyperdimensional computing (HDC). The feature extractor exploits the weight clustering mechanism to reduce computational complexity, while the HDC-based FSL classifier eliminates gradient-based back propagation operations, enabling single-pass training with substantially reduced latency. Additionally, FSL-HDnn enables low-latency ODL and inference via two proposed optimization strategies, including an early-exit mechanism with branch feature extraction and batched single-pass training that improves hardware utilization. Measurement results demonstrate that our chip fabricated in a 40 nm CMOS process delivers superior training energy efficiency of 6 mJ/image and end-to-end training throughput of 28 images/s on a 10-way 5-shot FSL task. The end-to-end training latency is also reduced by 2x to 20.9x compared to state-of-the-art ODL chips.
title FSL-HDnn: A 40 nm Few-shot On-Device Learning Accelerator with Integrated Feature Extraction and Hyperdimensional Computing
topic Hardware Architecture
Image and Video Processing
url https://arxiv.org/abs/2512.11826