{
  "id": "1604.07043",
  "version": "v1",
  "published": "2016-04-24T15:53:22.000Z",
  "updated": "2016-04-24T15:53:22.000Z",
  "title": "Towards Better Analysis of Deep Convolutional Neural Networks",
  "authors": [
    "Mengchen Liu",
    "Jiaxin Shi",
    "Zhen Li",
    "Chongxuan Li",
    "Jun Zhu",
    "Shixia Liu"
  ],
  "comment": "Submitted to VIS 2016",
  "categories": [
    "cs.CV"
  ],
  "abstract": "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.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2016-04-24T15:53:22.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "deep convolutional neural networks",
      "better analysis",
      "deep cnn",
      "high-quality deep models typically relies",
      "reduce visual clutter"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable",
      "adsabs": "2016arXiv160407043L"
    }
  }
}