On Transversality of Bent Hyperplane Arrangements and the Topological Expressiveness of ReLU Neural Networks.
J. Elisenda GrigsbyKathryn LindseyPublished in: SIAM J. Appl. Algebra Geom. (2022)
Keyphrases
- hyperplane
- neural network
- training samples
- support vector
- feature space
- input space
- data points
- support vector machine
- principal components
- linear classifiers
- pattern recognition
- maximal margin
- decision boundary
- artificial neural networks
- linearly separable
- incremental learning algorithm
- locality sensitive
- kernel function
- convex hull
- svm classifier
- back propagation
- linear separability
- classification procedure
- average distance
- weight vector
- normal vectors
- kernel space
- support vectors
- active learning