Thanks for your question. First the diagram is not meant to illustrate the
exact links between the hierarchies. It is meant to illustrate that control
binons (nodes) are needed between percepts and acts. One type of control binon
connects percepts to acts such that if the action habit is active and the
percept is recognized it triggers the act to be done. A second type of control
binon connects acts to percepts such that when an act is done the resulting
percept can be expected. The challenge in forming these structures and
controlling their performance is that given a trigger percept many acts may be
done and given an act there are many possible resulting stimuli that could be
expected. And given a resulting percept there may be many acts that may produce
it. And an act (like talking on the phone) must be reusable in many different
context situations. The control nodes provide the relational structure for the
forward and inverse models in motor control.
https://www.cs.bham.ac.uk/research/projects/cogaff/aiib/Symposium_6/Papers/Cooper.pdf
------------------------------------------
Artificial General Intelligence List: AGI
Permalink:
https://agi.topicbox.com/groups/agi/T243fc3536ee14d1d-M2bd93c563f91141cb94f374f
Delivery options: https://agi.topicbox.com/groups/agi/subscription
