Corpus ID: 15654042

On the Stability of Deep Networks

@article{Giryes2014OnTS,
  title={On the Stability of Deep Networks},
  author={Raja Giryes and Guillermo Sapiro and Alexander M. Bronstein},
  journal={CoRR},
  year={2014},
  volume={abs/1412.5896}
}
  • Raja Giryes, Guillermo Sapiro, Alexander M. Bronstein
  • Published in ICLR 2014
  • Mathematics, Computer Science
  • CoRR
  • In this work we study the properties of deep neural networks (DNN) with random weights. We formally prove that these networks perform a distance-preserving embedding of the data. Based on this we then draw conclusions on the size of the training data and the networks' structure. A longer version of this paper with more results and details can be found in (Giryes et al., 2015). In particular, we formally prove in the longer version that DNN with random Gaussian weights perform a distance… CONTINUE READING

    Create an AI-powered research feed to stay up to date with new papers like this posted to ArXiv

    Citations

    Publications citing this paper.
    SHOWING 1-3 OF 3 CITATIONS

    Unconstrained Still/Video-Based Face Verification with Deep Convolutional Neural Networks

    VIEW 1 EXCERPT
    CITES BACKGROUND

    Towards the design of an end-to-end automated system for image and video-based recognition

    VIEW 1 EXCERPT
    CITES BACKGROUND

    An End-to-End System for Unconstrained Face Verification with Deep Convolutional Neural Networks

    VIEW 1 EXCERPT

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 13 REFERENCES

    Dimension Reduction by Random Hyperplane Tessellations

    VIEW 3 EXCERPTS
    HIGHLY INFLUENTIAL

    Empirical processes and random projections

    VIEW 1 EXCERPT
    HIGHLY INFLUENTIAL

    Understanding deep image representations by inverting them

    VIEW 1 EXCERPT

    Signal recovery from Pooling Representations

    VIEW 1 EXCERPT