On 4/17/2019 6:00 PM, Russell Standish wrote:
On Wed, Apr 17, 2019 at 06:25:19PM +0200, Bruno Marchal wrote:Rover is conscious, but still dissociated from ‘rover”. But that is just because it has no strong induction axiom, and no way to build approximation of models of itself. It lack a re-entring neural system rich enough to manage the gap between its first person apprehension, and the third person apparent reality around it. The entity "Telmo" exists in your mind and mine, and I happen to be an entity "Telmo" in whose mind the entity "Telmo" also exists. This is real self-reference. I agree. It is unclear for me if Mars Rover has it, or not, as I have not seen the code, and even seeing it, it could ba a Helle of a difficulty to prove it has not that ability. I doubt it has it, because Naza does not want a free exploratory on Mars, but a docile slave.I do think self-reference has something to do with it, as without an observer to give meaning to something, it has no meaning. For instance, without an observer to interpret a certain pile of atoms as a machine, it is just a pile of atoms. Unless you propose a la Bishop Berkley some sort of devine mind from which all meaning radiates, the only other possibility is that each consciousness bootstraps its own meaning from self-reference. Unless the mars rover has a self model in its code (and I don't think it was constructed that way),
But how complete must the self-model be. As Bruno has pointed out, it can't be complete. Current Mars Rovers have some "house keeping"self-knowledge, like battery charge, temperature, power draw, next task, location, time,... Of course current rovers don't have AI which would entail them learning and planning, which would require that they be able to run a simulation which included some representation of themself; but that representation might be very simple. When you plan to travel to the next city your plan includes a representation of yourself, but probably only as a location.
Brent
then I would extremely doubt it has any sort of consciousness. A more interesting possibility is Hod Lipson's "starfish" robot, which has self-reference baked in. Cheers
-- You received this message because you are subscribed to the Google Groups "Everything List" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout.
