fcCovaddddddw: Covariance between first and second derivatives from...

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

Description

This function calculates covariance between weights and second derivatives and covariance between first and second derivatives from consecutive layers.

Usage

fcCovaddddddw(mao, mai, mdao, Caoai, Cdodi, Caow, Cdow, acto, acti, ni, no, B)

Arguments

mao	Mean vector of activation units from next layer
mai	Mean vector of activation units from current layer
mdao	Mean vector of activation units' first derivative from next layer
Caoai	Covariance between activation units from current and next layers
Cdodi	Covariance between first derivatives from current and next layers
Caow	Covariance between activation units and weights
Cdow	Covariance between derivatives and weights
acto	Activation function index for next layer defined by activationFunIndex
acti	Activation function index for current layer defined by activationFunIndex
ni	Number of units in current layer
no	Number of units in next layer
B	Batch size

Value

• Covariance between first and second derivatives from consecutive layers

• Covariance between second derivatives from next layer and weights

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