ENAS：面向资源受限微控制器上TinyML的高效硬件感知神经架构搜索框架

中文摘要 AI 辅助

我们提出\textbf{ENAS}，一个硬件感知的神经架构搜索（NAS）框架，它结合了静态可行性检查、支持标准块、深度可分离块和瓶颈块（可选跳跃连接）的基于细胞的搜索空间，以及一个三阶段混合搜索策略（随机$\rightarrow$ top-$K$ $\rightarrow$ 变异），并带有跨运行持久缓存。与许多依赖GPU加速的现有NAS框架不同，ENAS设计为无需GPU即可高效运行，使其适用于资源受限的开发环境。我们在两个TinyML基准（Visual Wake Words和Melanoma Cancer）上评估ENAS，涵盖八个微控制器，内存占用从20 KB到1 MB SRAM，以及九种输入图像分辨率。我们的实验结果表明，与最近的NanoNAS框架相比，ENAS在Visual Wake Words和Melanoma Cancer数据集上分别实现了平均搜索时间加速$2.41{\times}$和$1.70{\times}$，同时保持了具有竞争力的测试准确率。一项实测资源分析进一步表明，ENAS选择的模型在匹配准确率下使用显著更低的峰值激活RAM，这是微控制器部署的约束条件。此外，ENAS在基于STM32H743的微控制器上实现了$79.4\%$的测试准确率，比贪婪的仅CPU基线高出$2.6$个百分点。我们将ENAS框架作为开源发布，网址为：this https URL

英文摘要

We present \textbf{ENAS}, a hardware-aware Neural Architecture Search (NAS) framework that combines a static feasibility check, a cell-based search space supporting standard, depthwise-separable, and bottleneck blocks with optional skip connections, and a three-stage hybrid search strategy (random $\rightarrow$ top-$K$ $\rightarrow$ mutation) with persistent cross-run caching. Unlike many existing NAS frameworks that rely on GPU acceleration, ENAS is designed to operate efficiently without requiring GPUs, making it suitable for resource-constrained development environments. We evaluate ENAS on two TinyML benchmarks, Visual Wake Words and Melanoma Cancer, across eight microcontrollers with memory footprints ranging from 20\,KB to 1\,MB SRAM and nine input image resolutions. Our experimental results show that ENAS achieves mean search-time speedups of $2.41{\times}$ and $1.70{\times}$ on the Visual Wake Words and Melanoma Cancer datasets, respectively, while maintaining competitive test accuracy compared with the recent NanoNAS framework. A measured resource analysis further shows that ENAS-selected models use substantially lower peak activation RAM, the binding constraint for microcontroller deployment at matched accuracy. Additionally, ENAS achieves $79.4\%$ test accuracy on an STM32H743-based microcontroller, outperforming the greedy CPU-only baseline by $2.6$ percentage points. We release the ENAS framework as open-source at: https://github.com/EdgeIntelligenceLab/ENAS

发表机构

• Indian Institute of Science (IISc)（印度科学研究所）

机构由 AI 辅助整理，请以论文原文为准。