Predicting Adversarial Examples with High Confidence
arXiv:1802.04457
Abstract
It has been suggested that adversarial examples cause deep learning models to make incorrect predictions with high confidence. In this work, we take the opposite stance: an overly confident model is more likely to be vulnerable to adversarial examples. This work is one of the most proactive approaches taken to date, as we link robustness with non-calibrated model confidence on noisy images, providing a data-augmentation-free path forward. The adversarial examples phenomenon is most easily explained by the trend of increasing non-regularized model capacity, while the diversity and number of samples in common datasets has remained flat. Test accuracy has incorrectly been associated with true generalization performance, ignoring that training and test splits are often extremely similar in terms of the overall representation space. The transferability property of adversarial examples was previously used as evidence against overfitting arguments, a perceived random effect, but overfitting is not always random.
arXiv commentUnder review by the International Conference on Machine Learning (ICML)
References in corpus (18)
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Binarized Neural Networks: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1
- Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation
- DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients
- On Calibration of Modern Neural Networks
- Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examples
- Synthesizing Robust Adversarial Examples
- Towards Deep Neural Network Architectures Robust to Adversarial Examples
- The Space of Transferable Adversarial Examples
- Countering Adversarial Images using Input Transformations
- Adversarial Machine Learning at Scale
- PixelDefend: Leveraging Generative Models to Understand and Defend against Adversarial Examples
- Defensive Distillation is Not Robust to Adversarial Examples
- Stochastic Activation Pruning for Robust Adversarial Defense
- Measuring the tendency of CNNs to Learn Surface Statistical Regularities
- Parseval Networks: Improving Robustness to Adversarial Examples
- A Boundary Tilting Persepective on the Phenomenon of Adversarial Examples
- High Dimensional Spaces, Deep Learning and Adversarial Examples