Corpus ID: 6470548

What Happened to My Dog in That Network: Unraveling Top-down Generators in Convolutional Neural Networks

  title={What Happened to My Dog in That Network: Unraveling Top-down Generators in Convolutional Neural Networks},
  author={P. W. Gallagher and Shuai Tang and Zhuowen Tu},
  • P. W. Gallagher, Shuai Tang, Zhuowen Tu
  • Published 2015
  • Computer Science, Mathematics
  • ArXiv
  • Top-down information plays a central role in human perception, but plays relatively little role in many current state-of-the-art deep networks, such as Convolutional Neural Networks (CNNs). This work seeks to explore a path by which top-down information can have a direct impact within current deep networks. We explore this path by learning and using "generators" corresponding to the network internal effects of three types of transformation (each a restriction of a general affine transformation… CONTINUE READING
    5 Citations

    Figures, Tables, and Topics from this paper

    Controllable Top-down Feature Transformer
    • 1
    • Highly Influenced
    • PDF
    Top-down Flow Transformer Networks
    • 1
    Mining Interpretable AOG Representations from Convolutional Networks via Active Question Answering
    • 1
    • PDF
    Multi-Shot Mining Semantic Part Concepts in CNNs
    • 1
    • PDF


    Deep neural networks are easily fooled: High confidence predictions for unrecognizable images
    • 1,840
    • PDF
    Intriguing properties of neural networks
    • 6,100
    • PDF
    Generative Modeling of Convolutional Neural Networks
    • 62
    • PDF
    Understanding deep image representations by inverting them
    • 1,185
    • PDF
    Inverting Visual Representations with Convolutional Networks
    • A. Dosovitskiy, T. Brox
    • Computer Science
    • 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
    • 2016
    • 365
    • PDF
    Spatial Transformer Networks
    • 3,376
    • PDF
    Inverting Convolutional Networks with Convolutional Networks
    • 95
    • Highly Influential
    • PDF
    ImageNet classification with deep convolutional neural networks
    • 59,736
    • PDF
    Sequence to Sequence Learning with Neural Networks
    • 11,707
    • PDF
    Unsupervised Learning of Image Transformations
    • 203
    • PDF