projUNN: efficient method for training deep networks with unitary matrices.
Bobak Toussi KianiRandall BalestrieroYann LeCunSeth LloydPublished in: CoRR (2022)
Keyphrases
- experimental evaluation
- high accuracy
- support vector machine svm
- pairwise
- network model
- highly efficient
- dynamic programming
- high precision
- synthetic data
- detection method
- computationally efficient
- support vector machine
- cost function
- feed forward neural networks
- feature set
- training algorithm
- high efficiency
- training phase
- pairwise comparison
- missing data
- neural network
- k means
- artificial neural networks
- preprocessing
- computational complexity
- objective function
- decision trees