Classifiers Based on Deep Sparse Coding Architectures are Robust to Deep Learning Transferable Examples.
Jacob M. SpringerCharles M. S. StraussAustin M. ThresherEdward KimGarrett T. KenyonPublished in: CoRR (2018)
Keyphrases
- deep learning
- sparse coding
- unsupervised learning
- deep belief networks
- dictionary learning
- unsupervised feature learning
- training examples
- sparse representation
- natural images
- image classification
- supervised learning
- machine learning
- linear combination
- restricted boltzmann machine
- generative model
- feature selection
- training data
- object recognition
- decision trees
- image representation
- dimensionality reduction
- semi supervised
- mental models
- learning algorithm
- feature space
- hierarchical models
- image patches
- partial occlusion
- active learning
- weakly supervised
- support vector
- data mining