arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2011.02955cs.LGcs.NEcs.SD
收藏

基于感受场正则化和频率阻尼的低复杂度声场景分类模型

Low-Complexity Models for Acoustic Scene Classification Based on Receptive Field Regularization and Frequency Damping

Johannes Kepler University Linz(林茨约翰内斯·开普勒大学)

Khaled Koutini, Florian Henkel, Hamid Eghbal-zadeh, Gerhard Widmer

摘要:

众所周知,深度神经网络在计算和内存要求方面非常苛刻。由于资源预算有限的嵌入式系统和移动设备的使用越来越多,设计低复杂度模型而不牺牲太多的预测性能变得非常重要。在这项工作中,我们调查和比较几个著名的方法,以减少神经网络中的参数的数量。我们进一步将这些放入最近关于感受场(RF)对模型性能影响的研究中,并根据经验表明,我们可以通过对RF施加特定限制,结合参数简化方法来实现高性能低复杂度模型。此外,我们提出了一种滤波器阻尼技术,用于正则化模型的RF,而不改变它们的架构和参数计数。我们将证明,采用这种技术可以提高各种低复杂度设置(如修剪和分解卷积)的性能。使用我们提出的滤波器阻尼,我们在DCASE-2020挑战赛中实现了低复杂度声学场景分类任务的第一名。

英文摘要:

Deep Neural Networks are known to be very demanding in terms of computing and memory requirements. Due to the ever increasing use of embedded systems and mobile devices with a limited resource budget, designing low-complexity models without sacrificing too much of their predictive performance gained great importance. In this work, we investigate and compare several well-known methods to reduce the number of parameters in neural networks. We further put these into the context of a recent study on the effect of the Receptive Field (RF) on a model's performance, and empirically show that we can achieve high-performing low-complexity models by applying specific restrictions on the RFs, in combination with parameter reduction methods. Additionally, we propose a filter-damping technique for regularizing the RF of models, without altering their architecture and changing their parameter counts. We will show that incorporating this technique improves the performance in various low-complexity settings such as pruning and decomposed convolution. Using our proposed filter damping, we achieved the 1st rank at the DCASE-2020 Challenge in the task of Low-Complexity Acoustic Scene Classification.

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。

↑