Feature Purification: How Adversarial Training Performs Robust Deep Learning.
Zeyuan Allen-ZhuYuanzhi LiPublished in: CoRR (2020)
Keyphrases
- deep learning
- deep architectures
- restricted boltzmann machine
- machine learning
- unsupervised learning
- unsupervised feature learning
- training set
- image features
- online learning
- feature set
- mental models
- training samples
- generative model
- semi supervised
- probabilistic model
- feature vectors
- viewpoint
- bayesian networks
- image segmentation
- feature selection