Pareto Optimization of CNN Models via Hardware-Aware Neural Architecture Search for Drainage Crossing Classification on Resource-Limited Devices
Pareto Optimization of CNN Models via Hardware-Aware Neural Architecture Search for Drainage Crossing Classification on Resource-Limited Devices
复制标题
DOI：
10.1145/3624062.3624258
复制
发表时间：
2023-11
期刊：
影响因子：
--
通讯作者：
Yuke Li;Jiwon Baik;Md Marufi Rahman;Iraklis Anagnostopoulos;Ruopu Li;Tong Shu
中科院分区：
文献类型：
--
作者：
Yuke Li;Jiwon Baik;Md Marufi Rahman;Iraklis Anagnostopoulos;Ruopu Li;Tong Shu
嵌入式设备受到有限的内存和处理器的限制，需要根据其规格定制深度学习模型。本研究探索了用于分类排水交叉图像的定制模型架构。本文以ResNet-18为基础，旨在最大限度地提高预测精度，减少内存大小，并最大限度地减少推理延迟。通过利用硬件感知神经架构搜索，系统地探测了各种配置，在六个基准测试变体中积累了1,717个实验结果。通过n- meter增强的实验数据分析，提供了对四种不同预测因子的推理延迟的全面理解。值得注意的是，具有准确性，延迟和记忆三个目标的帕累托前分析产生了五个非主导解决方案。这些杰出的模型在保持精度的同时显示了效率，当部署在资源受限的环境中时，为传统的ResNet-18提供了令人信服的替代方案。论文最后强调了从结果中得出的见解，并提出了未来探索的途径。
Embedded devices, constrained by limited memory and processors, require deep learning models to be tailored to their specifications. This research explores customized model architectures for classifying drainage crossing images. Building on the foundational ResNet-18, this paper aims to maximize prediction accuracy, reduce memory size, and minimize inference latency. Various configurations were systematically probed by leveraging hardware-aware neural architecture search, accumulating 1,717 experimental results over six benchmarking variants. The experimental data analysis, enhanced by nn-Meter, provided a comprehensive understanding of inference latency across four different predictors. Significantly, a Pareto front analysis with three objectives of accuracy, latency, and memory resulted in five non-dominated solutions. These standout models showcased efficiency while retaining accuracy, offering a compelling alternative to the conventional ResNet-18 when deployed in resource-constrained environments. The paper concludes by highlighting insights drawn from the results and suggesting avenues for future exploration.