An unfolding method based on conditional Invertible Neural Networks (cINN) using iterative training.
Mathias BackesAnja ButterMonica DunfordBogdan MalaescuPublished in: CoRR (2022)
Keyphrases
- training process
- detection method
- similarity measure
- neural network
- computational cost
- high accuracy
- data sets
- dynamic programming
- significant improvement
- cost function
- preprocessing
- experimental evaluation
- computational complexity
- training algorithm
- pattern recognition
- pairwise
- video sequences
- support vector machine svm
- feedforward neural networks
- feed forward neural networks
- feature extraction
- high precision
- fault diagnosis
- synthetic data
- clustering method
- fuzzy logic
- support vector machine
- artificial neural networks