arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2307.09488cs.LG
收藏

PLiNIO: 面向复杂度感知DNN优化的易用梯度方法库

PLiNIO: A User-Friendly Library of Gradient-based Methods for Complexity-aware DNN Optimization

Politecnico di Torino(都灵理工大学)

Daniele Jahier Pagliari, Matteo Risso, Beatrice Alessandra Motetti, Alessio Burrello

AI总结:

本研究提出开源库PLiNIO,在统一易用界面下集成轻量级梯度驱动的DNN设计自动化技术,经边缘任务实验验证,其组合优化方案在准确率-模型尺寸权衡上帕累托优于基准,最高可减94.34%内存且准确率降不足1%。

摘要:

精准且高效的深度神经网络(DNN)需求旺盛,尤其适用于需在资源受限边缘设备上运行的应用场景。由于超参数组合空间极为庞大,无法通过人工全面探索,为新应用在合理时间内找到此类DNN需要自动化优化流程。本研究提出PLiNIO——一款开源库,在统一且易用的界面下实现了一整套最先进的DNN设计自动化技术,所有技术均基于轻量级梯度优化。通过在多项边缘相关任务上开展实验,我们证明结合PLiNIO提供的各类优化方法,可得到丰富的解集,这些解集在准确率与模型尺寸的权衡上帕累托优于所选取的基准。值得注意的是,与基准架构相比,PLiNIO可实现最高94.34%的内存缩减,而准确率下降不足1%。

英文摘要:

Accurate yet efficient Deep Neural Networks (DNNs) are in high demand, especially for applications that require their execution on constrained edge devices. Finding such DNNs in a reasonable time for new applications requires automated optimization pipelines since the huge space of hyper-parameter combinations is impossible to explore extensively by hand. In this work, we propose PLiNIO, an open-source library implementing a comprehensive set of state-of-the-art DNN design automation techniques, all based on lightweight gradient-based optimization, under a unified and user-friendly interface. With experiments on several edge-relevant tasks, we show that combining the various optimizations available in PLiNIO leads to rich sets of solutions that Pareto-dominate the considered baselines in terms of accuracy vs model size. Noteworthy, PLiNIO achieves up to 94.34% memory reduction for a <1% accuracy drop compared to a baseline architecture.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑