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 2107.14062cs.LGcs.AIcs.CVphysics.app-phphysics.comp-ph
收藏

全连接神经网络的结构与性能:新兴的复杂网络特性

Structure and Performance of Fully Connected Neural Networks: Emerging Complex Network Properties

São Carlos Institute of Physics, University of São Paulo(圣保罗大学圣卡洛斯物理研究所)

Leonardo F. S. Scabini, Odemir M. Bruno

摘要:

理解人工神经网络的行为是最近该领域的主要课题之一,因为自从深度学习广泛传播以来,黑盒方法已经变得很常见。这样的高维模型可能会表现出类似复杂系统的不稳定性和怪异特性。因此,我们提出了复杂网络(CN)技术来分析全连接神经网络的结构和性能。为此,我们构建了一个包含4000个模型及其各自CN属性的数据集。他们采用监督分类设置考虑四个视觉基准。每个神经网络被视为神经元和突触的加权无向图,并在训练后计算中心性度量。结果表明,这些措施是高度相关的网络分类性能。我们还提出了Bag-Of-Neurons(BoN)的概念,这是一种基于CN的方法,用于寻找连接相似神经元的拓扑签名。结果表明,六种神经元类型出现在这样的网络中,独立于目标域,并根据分类精度分布不同。我们还解决了与性能相关的特定CN属性,例如在性能较低的模型上具有较高的子图中心性。我们的研究结果表明,CN属性在全连接神经网络的性能中起着至关重要的作用,拓扑模式独立出现在各种模型中。

英文摘要:

Understanding the behavior of Artificial Neural Networks is one of the main topics in the field recently, as black-box approaches have become usual since the widespread of deep learning. Such high-dimensional models may manifest instabilities and weird properties that resemble complex systems. Therefore, we propose Complex Network (CN) techniques to analyze the structure and performance of fully connected neural networks. For that, we build a dataset with 4 thousand models and their respective CN properties. They are employed in a supervised classification setup considering four vision benchmarks. Each neural network is approached as a weighted and undirected graph of neurons and synapses, and centrality measures are computed after training. Results show that these measures are highly related to the network classification performance. We also propose the concept of Bag-Of-Neurons (BoN), a CN-based approach for finding topological signatures linking similar neurons. Results suggest that six neuronal types emerge in such networks, independently of the target domain, and are distributed differently according to classification accuracy. We also tackle specific CN properties related to performance, such as higher subgraph centrality on lower-performing models. Our findings suggest that CN properties play a critical role in the performance of fully connected neural networks, with topological patterns emerging independently on a wide range of models.

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。

↑