Variable outputs can be described by rules
Even granting that a given state of a connectionist network could lead to a variety of further states, a set of rules could still be formulated to describe its behaviour.
This is because rule forms have:
 
"implicit conventions of simplification when a single input representation might lead to distinct output representations".

So, the syntactic argument holds.

Kenneth Aizawa, 1994, p.484.


Note
: The debate between Aizawa and Horgan & Tienson about rules is carried out with much more precision than could be captured here. In the technical discussion, issues of quasi-exceptionless rules, probabalistic laws, and ceteris paribus rules are raised.
CONTEXT(Help)
-
Artificial Intelligence »Artificial Intelligence
Can computers think? [1] »Can computers think? [1]
Yes: connectionist networks can think [5a] »Yes: connectionist networks can think [5a]
Connectionist networks can think without following rules »Connectionist networks can think without following rules
Representations without rules »Representations without rules
The Syntactic Argument »The Syntactic Argument
The Multiple Realisability Defence  »The Multiple Realisability Defence
Variable outputs can be described by rules
+Comments (0)
+Citations (0)
+About