Metamorphic detection of adversarial examples in deep learning models with affine transformations.
Rohan Reddy MekalaGudjon Einar MagnussonAdam PorterMikael LindvallMadeline DiepPublished in: MET@ICSE (2019)
Keyphrases
- learning models
- affine transformation
- affine invariant
- learning algorithm
- machine learning
- semi supervised learning
- learning tasks
- loss function
- image registration
- image mosaicking
- feature points
- machine learning algorithms
- b spline
- image matching
- similarity transformations
- training examples
- conditional random fields
- learning problems
- active learning
- image processing
- support vector
- training set
- classification models
- hidden markov models
- semi supervised
- text categorization
- markov random field