AtomNet: Designing Tiny Models from Operators Under Extreme MCU Constraints
Open Access
- 11 April 2025
- journal article
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 39 (15) , 16354-16362
Abstract
Tiny machine learning (TinyML) has attracted heightened attention for its ability to provide low-cost and instantaneous performance on edge devices. Particularly, the commonly used microcontroller unit (MCU) imposes extreme constraints on peak memory (SRAM) and storage (Flash). Existing TinyML methods often rely on a customized and hard-to-obtain inference libraries, as well as necessitate a time-consuming search for a deployable architecture using advanced Neural Architecture Search (NAS) algorithms. To solve these problems, we fully exploit the resources on MCU and deduce hardware-oriented guidelines for designing models under extreme MCU constraints. In detail, we delve into thorough information about the atom operators by collecting the runtime data of Flash, SRAM, and latency to build a dataset named AtomDB. Based on AtomDB, several critical operator guidelines are established to fully utilize limited Flash and SRAM, while minimizing latency. By transferring the guidelines to analyze blocks, we propose a hybrid pattern that organizes appropriate blocks at different network stages to form the AtomNet, a more hardware-oriented architecture, to handle the former SRAM bottleneck and the latter Flash bottleneck. Extensive experiments demonstrate the effectiveness of the exploitation of the hardware characteristics. Remarkably, AtomNet pioneeringly achieve 3.5% accuracy enhancement and more than 15% latency reduction on 320KB MCU using readily available official inference libraries for ImageNet tasks, surpassing the current state-of-the-art method.Keywords
This publication has 10 references indexed in Scilit:
- MobileOne: An Improved One millisecond Mobile Backbone2023
- NAS-Bench-Suite-Zero: Accelerating Research on Zero Cost Proxies2022
- Swin Transformer: Hierarchical Vision Transformer using Shifted Windows2021
- Searching for MobileNetV32019
- FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Search2019
- MnasNet: Platform-Aware Neural Architecture Search for Mobile2019
- AMC: AutoML for Model Compression and Acceleration on Mobile Devices2018
- MobileNetV2: Inverted Residuals and Linear Bottlenecks2018
- Deep Residual Learning for Image Recognition2016
- The Pascal Visual Object Classes (VOC) ChallengeInternational Journal of Computer Vision, 2009