arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.05397eess.AScs.LGcs.SD
收藏

面向节能常开音频模型的神经架构搜索

Neural Architecture Search for Energy Efficient Always-on Audio Models

X, The Moonshot Factory(X,登月工厂);Google Research(谷歌研究院)

Daniel T. Speckhard, Karolis Misiunas, Sagi Perel, Tenghui Zhu, Simon Carlile, Malcolm Slaney

AI总结:

针对常开音频分类的能耗与内存约束,提出同时优化准确率、能耗和内存的 NAS 策略,在 AudioSet 上获得十倍级推理节能并小幅提升准确率。

摘要:

用于常开分类任务的移动和边缘计算设备需要节能的神经网络架构。本文提出对神经架构搜索(NAS)的若干修改,以提高其在实际场景中的成功概率。我们的搜索同时优化网络准确率、能效和内存使用。我们在真实硬件上对搜索性能进行基准测试,但由于难以在真实硬件上运行数千次测试,因此使用随机森林模型粗略预测候选网络的能耗。我们提出一种搜索策略,结合使用贝叶斯搜索与带粒子群的正则化进化搜索,并采用早停机制降低计算负担。在基于 AudioSet 的声音事件分类数据集上评估时,与我们的基线 MobileNetV1/V2 实现相比,该搜索使每次推理的能耗降低一个数量级,内存占用大幅减小,同时任务准确率略有提升。我们还证明,将二维频谱图与具有大量滤波器的卷积结合会造成音频分类的计算瓶颈,而替代方法虽能减轻计算负担,却会牺牲任务准确率。

英文摘要:

Mobile and edge computing devices for always-on classification tasks require energy-efficient neural network architectures. In this paper we present several changes to neural architecture searches (NAS) that improve the chance of success in practical situations. Our search simultaneously optimizes for network accuracy, energy efficiency and memory usage. We benchmark the performance of our search on real hardware, but since running thousands of tests with real hardware is difficult we use a random forest model to roughly predict the energy usage of a candidate network. We present a search strategy that uses both Bayesian and regularized evolutionary search with particle swarms, and employs early-stopping to reduce the computational burden. Our search, evaluated on a sound-event classification dataset based upon AudioSet, results in an order of magnitude less energy per inference and a much smaller memory footprint than our baseline MobileNetV1/V2 implementations while slightly improving task accuracy. We also demonstrate how combining a 2D spectrogram with a convolution with many filters causes a computational bottleneck for audio classification and that alternative approaches reduce the computational burden but sacrifice task accuracy.

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。

↑