Implicit Bias of Gradient Descent for Mean Squared Error Regression with Two-Layer Wide Neural Networks.
Hui JinGuido MontúfarPublished in: J. Mach. Learn. Res. (2023)
Keyphrases
- neural network
- multi layer
- regression model
- learning rules
- single layer
- cost function
- pattern recognition
- neural nets
- linear regression
- artificial neural networks
- back propagation
- support vector
- support vector regression
- neural network model
- fault diagnosis
- prediction error
- regression analysis
- feedforward neural networks
- regression algorithm
- ridge regression
- locally weighted
- fuzzy systems
- multilayer perceptron
- fuzzy logic
- training process
- network architecture
- model selection
- regression problems
- partial least squares
- least squares
- simple linear
- feature space
- decision trees
- genetic algorithm