fcCombinaisonDweightNode: Combination of squared products of first derivative

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

Description

This function calculates mean of squared products of first derivatives (wd)^2. Every products (weight times node) from current layer are considered which results in a (B*ni x no)-matrix.

Usage

fcCombinaisonDweightNode(mpdi, mw, Sw, mda, Sda, ni, no, B)

Arguments

mpdi	Mean vector of first derivative product wd of current layer
mw	Mean vector of weights for the current layer
Sw	Covariance of weights for the current layer
mda	Mean vector of activation units' first derivative from current layer
Sda	Covariance of activation units' first derivative from current layer
ni	Number of units in current layer
no	Number of units in next layer
B	Batch size

Value

Mean matrix of squared products of first derivatives

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