... neurons

When referring to 'neurons', the computational neurobiological interpretation is assumed.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... Lappalainen

The former name of the present author was Lappalainen.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... propositions

Since fuzzy logic is sometimes presented as an alternative to Bayesian probability theory, it should be emphasised that truth value and degree of belief are different dimensions. It is possible to envision an extension of Bayesian probability theory which would define probabilities for propositions with fuzzy truth values.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... factors

The factors are also termed hidden causes, latent variables or sources in the literature.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... layers

There are two conventions for counting the layers. Here the factors are counted as one layer although they are not computational elements. The alternative is to count only layers of computational elements. Both conventions agree on how to count hidden layers.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... V

Matlab code for running the simulations is available at http://www.cis.hut.fi/projects/ica/bayes/.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.