An initial alignment between neural network and target is needed for gradient descent to learn.
Emmanuel AbbeElisabetta CornacchiaJan HazlaChristopher MarquisPublished in: CoRR (2022)
Keyphrases
- neural network
- learning rules
- pattern recognition
- genetic algorithm
- pre trained
- cost function
- procrustes analysis
- target tracking
- feed forward neural networks
- self organizing maps
- fuzzy logic
- artificial neural networks
- feedforward neural networks
- error function
- sequence alignment
- neural model
- neural nets
- fuzzy neural network
- learning rate
- multi layer
- protein sequences
- back propagation
- knn
- objective function
- similarity measure
- machine learning