arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.04583cs.LGcs.AI
收藏

揭示稀疏神经网络在特征选择中的能力

Unveiling the Power of Sparse Neural Networks for Feature Selection

University of Twente(特温特大学);University of Cambridge(剑桥大学);Eindhoven University of Technology(埃因霍温理工大学);University of Luxembourg(卢森堡大学)

Zahra Atashgahi, Tennison Liu, Mykola Pechenizkiy, Raymond Veldhuis, Decebal Constantin Mocanu, Mihaela van der Schaar

AI总结:

本文系统分析了基于动态稀疏训练的稀疏神经网络在特征选择中的表现,并提出一种新的特征重要性度量指标,证明其能在显著降低内存和计算开销的同时提升所选特征质量。

摘要:

稀疏神经网络(SNNs)已成为高效特征选择的有力工具。利用SNNs中的动态稀疏训练(DST)算法已展现出良好的特征选择能力,同时大幅降低了计算开销。尽管取得了这些进展,但在特征选择方面仍有几个关键方面未被充分探索。关于网络训练中DST算法的选择、用于对特征/神经元进行排序的度量指标的选择,以及与密集网络相比这些方法在不同数据集上的比较性能等问题仍然存在。本文通过对稀疏神经网络特征选择的全面系统分析来填补这些空白。此外,我们引入了一种考虑稀疏神经网络特性的新度量指标,旨在量化SNNs背景下特征的重要性。我们的研究结果表明,与密集网络相比,使用DST算法训练的SNNs进行特征选择平均可实现超过50%的内存减少和55%的FLOPs减少,同时在所选特征的质量方面优于密集网络。我们的代码和补充材料可在GitHub上获取(\url{https://github.com/zahraatashgahi/Neuron-Attribution})。

英文摘要:

Sparse Neural Networks (SNNs) have emerged as powerful tools for efficient feature selection. Leveraging the dynamic sparse training (DST) algorithms within SNNs has demonstrated promising feature selection capabilities while drastically reducing computational overheads. Despite these advancements, several critical aspects remain insufficiently explored for feature selection. Questions persist regarding the choice of the DST algorithm for network training, the choice of metric for ranking features/neurons, and the comparative performance of these methods across diverse datasets when compared to dense networks. This paper addresses these gaps by presenting a comprehensive systematic analysis of feature selection with sparse neural networks. Moreover, we introduce a novel metric considering sparse neural network characteristics, which is designed to quantify feature importance within the context of SNNs. Our findings show that feature selection with SNNs trained with DST algorithms can achieve, on average, more than $50\%$ memory and $55\%$ FLOPs reduction compared to the dense networks, while outperforming them in terms of the quality of the selected features. Our code and the supplementary material are available on GitHub (\url{https://github.com/zahraatashgahi/Neuron-Attribution}).

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。

↑