跳到正文
arXiv cs.AI· Guilherme Guedes, Marco Apolinario Lainez, Charlotte Frenkel·· 10 小时前AI 评分31

Hypnos:面向边缘端可扩展持续学习的双阶段端到端平台

Hypnos: A Dual-Phase End-to-End Platform for Scalable Continual Learning at the Edge

AI 导读

Hypnos 是一个开源双阶段边缘持续学习平台,模仿大脑的清醒与睡眠周期:清醒阶段在 RISC-V MCU 系统级芯片上做无梯度原型更新,睡眠阶段由智能充电器执行基于回放的梯度更新以巩固知识。

正文 · 原文

View PDF HTML (experimental)

Abstract:Continual learning (CL) enables neural networks to adapt to non-stationary data streams by mitigating catastrophic forgetting, offering a path to robust, private, and adaptable edge devices. Yet, porting CL to real-world edge devices remains challenging for two reasons: (i) online operation, where sensor data arrives one sample at a time and the device must adapt on-the-fly, and (ii) the low footprint of edge devices, which makes traditional error backpropagation training prohibitive. Inspired by the brain's wake and sleep cycles, the former for rapid adaptation with minimal resources and the latter for long-term consolidation, we introduce Hypnos, an open-source dual-phase platform for efficient CL at the edge. In the wake phase (e.g., daily operation), Hypnos relies on a low-footprint system-on-chip with a RISC-V microcontroller unit (MCU) targeting efficient inference and supporting lightweight gradient-free prototype updates. In the sleep phase (e.g., night-time charging), where wall-plug energy and larger memory budgets are available, a smart charger performs rehearsal-based gradient updates for consolidation. Deployed on a ZCU104 field-programmable gate array (FPGA) and synthesized in a 28-nm process, Hypnos is benchmarked on the CORe50 NICv2-391 protocol. It sets a new state-of-the-art accuracy of 78.6%-81.6% for edge CL systems, with an on-device latency of 7.79s-10.54s per sample depending on activation precision, and requires $3.8\times$ fewer learning steps and $22\times$ less learning compute than quantized latent replay at higher accuracy and comparable latency.
Subjects: Hardware Architecture (cs.AR); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
Cite as: arXiv:2611.24637 [cs.AR]
  (or arXiv:2611.24637v1 [cs.AR] for this version)
  https://doi.org/10.48550/arXiv.2611.24637

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Guilherme Guedes [view email]
[v1] Wed, 18 Nov 2026 15:46:58 UTC (1,541 KB)

来源:arXiv cs.AI · arxiv.org