fcCwdowdowwdi2_3hl: Covariance between next layer multiplied products and current...

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

Description

This function calculates covariance cov(wdowdo,wdiwdi) where the di terms are the same when next second layer involves multiplied terms (wdo2wdo2). It is used when there are at least 3 hidden layers.

Usage

fcCwdowdowwdi2_3hl(mpdi, mpdo, mdgo2, Cdgodgi, acti, ni, no, no2, B)

Arguments

mpdi	Mean vector of first derivative product wd of current layer
mpdo	Mean vector of first derivative product wd of next layer
mdgo2	Mean vector of product of derivatives in second next layer
Cdgodgi	Covariance between weights times derivatives from consecutive layers
acti	Activation function index for current layer defined by activationFunIndex
ni	Number of units in current layer
no	Number of units in next layer
no2	Number of units in second next layer
B	Batch size

Value

Covariance cov(wdowdo,wdiwdi) where the di terms are the same

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