Advancing neural network calibration: The role of gradient decay in large-margin Softmax optimization.
Siyuan ZhangLinbo XiePublished in: Neural Networks (2024)
Keyphrases
- neural network
- artificial neural networks
- activation function
- pattern recognition
- optimization problems
- optimization algorithm
- back propagation
- recurrent neural networks
- image reconstruction from projections
- gradient vector
- neural nets
- global optimization
- self organizing maps
- support vector
- constrained optimization
- focal length
- intrinsic parameters
- learning algorithm
- genetic algorithm
- optimization method
- steepest descent method
- neural network is trained
- highly non linear
- temporal difference learning
- nonlinear optimization
- power law
- training algorithm
- knn
- fuzzy logic