feedForward: Forward uncertainty propagation

In mgoulet847/tagi: Tractable Approximate Gaussian Inference in Neural Networks

Description

This function feeds the neural network forward from input data to responses.

Usage

feedForward(NN, x, mp, Sp)

Arguments

NN	Lists the structure of the neural network
x	Input data
mp	Mean vectors of parameters for each layer μ_{θ}
Sp	Covariance matrices of parameters for each layer Σ_{θ}

Value

• Mean vectors of units for each layer μ_{Z}

• Covariance matrices of units for each layer Σ_{Z}

• Covariance matrices between units and weights for each layer Σ_{ZW}

• Covariance matrices between units and biases for each layer Σ_{ZB}

• Covariance matrices between previous and current units for each layer Σ_{ZZ^{+}}

mgoulet847/tagi documentation built on Dec. 21, 2021, 5:10 p.m.