Adversarial Deep Learning Models with Multiple Adversaries

@article{Chivukula2019AdversarialDL,
  title={Adversarial Deep Learning Models with Multiple Adversaries},
  author={Aneesh Sreevallabh Chivukula and W. Liu},
  journal={IEEE Transactions on Knowledge and Data Engineering},
  year={2019},
  volume={31},
  pages={1066-1079}
}
We develop an adversarial learning algorithm for supervised classification in general and Convolutional Neural Networks (CNN) in particular. The algorithm's objective is to produce small changes to the data distribution defined over positive and negative class labels so that the resulting data distribution is misclassified by the CNN. The theoretical goal is to determine a manipulating change on the input data that finds learner decision boundaries where many positive labels become negative… Expand
3 Citations
RazorNet: Adversarial Training and Noise Training on a Deep Neural Network Fooled by a Shallow Neural Network
  • 4
  • PDF

References

SHOWING 1-10 OF 26 REFERENCES
Sparse Feature Attacks in Adversarial Learning
  • 21
Towards Deep Neural Network Architectures Robust to Adversarial Examples
  • 514
  • PDF
Generative Adversarial Nets
  • 21,899
  • Highly Influential
  • PDF
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
  • 7,695
  • PDF
Classifier evaluation and attribute selection against active adversaries
  • 67
  • PDF
Support vector machines under adversarial label contamination
  • 129
  • PDF
BEGAN: Boundary Equilibrium Generative Adversarial Networks
  • 822
  • PDF
InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
  • 2,408
  • PDF
Improved Training of Wasserstein GANs
  • 4,152
  • PDF
...
1
2
3
...