Corpus ID: 201668754

Deep Neural Network Ensembles against Deception: Ensemble Diversity, Accuracy and Robustness

@article{Liu2019DeepNN,
  title={Deep Neural Network Ensembles against Deception: Ensemble Diversity, Accuracy and Robustness},
  author={Ling Liu and Wenqi Wei and Ka-Ho Chow and Margaret Loper and Emre Gursoy and Stacey Truex and Yanzhao Wu},
  journal={ArXiv},
  year={2019},
  volume={abs/1908.11091}
}
  • Ling Liu, Wenqi Wei, +4 authors Yanzhao Wu
  • Published in ArXiv 2019
  • Mathematics, Computer Science
  • Ensemble learning is a methodology that integrates multiple DNN learners for improving prediction performance of individual learners. Diversity is greater when the errors of the ensemble prediction is more uniformly distributed. Greater diversity is highly correlated with the increase in ensemble accuracy. Another attractive property of diversity optimized ensemble learning is its robustness against deception: an adversarial perturbation attack can mislead one DNN model to misclassify but may… CONTINUE READING

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

    Tables and Topics from this paper.

    Explore Further: Topics Discussed in This Paper

    Citations

    Publications citing this paper.

    Cross-Layer Strategic Ensemble Defense Against Adversarial Examples

    VIEW 1 EXCERPT
    CITES BACKGROUND

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 36 REFERENCES

    Strategic Teaming of Input Ensemble and Output Ensemble

    • Wenqi Wei, Ling Liu
    • Against Deception",
    • 2019
    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    The Limitations of Deep Learning in Adversarial Settings

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Explaining and Harnessing Adversarial Examples

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Diversity creation methods: a survey and categorisation

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    Statistical methods for rates and proportions

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    Denoising and Verification Cross-Layer Ensemble Against Black-box Adversarial Attacks

    VIEW 1 EXCERPT