Towards Better Analysis of Deep Convolutional Neural Networks
Towards Better Analysis of Deep Convolutional Neural Networks
复制标题
更好地分析深度卷积神经网络
DOI：
10.1109/tvcg.2016.2598831
复制
发表时间：
2017-01-01
影响因子：
5.2
通讯作者：
Liu, Shixia
中科院分区：
文献类型：
--
作者：
Liu, Mengchen;Shi, Jiaxin;Liu, Shixia
深度卷积神经网络（cnn）在图像分类等许多模式识别任务中取得了突破性的表现。然而，高质量深度模型的开发通常依赖于大量的试错，因为对于深度模型何时以及为什么起作用仍然没有明确的理解。在本文中，我们提出了一种可视化分析方法，用于更好地理解、诊断和精炼深度cnn。我们将深度CNN表述为一个有向无环图。在此基础上，开发了一种混合可视化来揭示每个神经元的多个方面以及它们之间的相互作用。特别地，我们引入了一种分层矩形填充算法和一种矩阵重排序算法来显示神经元簇的派生特征。我们还提出了一种基于双聚类的边缘捆绑方法，以减少神经元之间大量连接造成的视觉杂波。我们在一组cnn上评估了我们的方法，结果总体上是有利的。
Deep convolutional neural networks (CNNs) have achieved breakthrough performance in many pattern recognition tasks such as image classification. However, the development of high-quality deep models typically relies on a substantial amount of trial-and-error, as there is still no clear understanding of when and why a deep model works. In this paper, we present a visual analytics approach for better understanding, diagnosing, and refining deep CNNs. We formulate a deep CNN as a directed acyclic graph. Based on this formulation, a hybrid visualization is developed to disclose the multiple facets of each neuron and the interactions between them. In particular, we introduce a hierarchical rectangle packing algorithm and a matrix reordering algorithm to show the derived features of a neuron cluster. We also propose a biclustering-based edge bundling method to reduce visual clutter caused by a large number of connections between neurons. We evaluated our method on a set of CNNs and the results are generally favorable.