On Thu, Jan 14, 2021 at 12:10 PM John Rose wrote:
>
> I noticed you pulled out the Sheldrake reference.
It just became irrelevant to the new simplified way of mapping CD
p-bits into PLN simple TVs ...
--
Artificial General Intelligence List: AGI
Permalink:
Good idea. The p-bits is clever but does open up a whole thing on emulation. I
noticed you pulled out the Sheldrake reference.
One of the reasons this paper is so good is that it relates to an actual AGI
system in advanced development, the OpenCog framework. And then all the logics
intermappe
> Look, if we don't build AGI, we are going to die.
Hurry up then!
--
Artificial General Intelligence List: AGI
Permalink:
https://agi.topicbox.com/groups/agi/T86b555c591599ac6-M337361d887d665af87a4fb91
Delivery options: https://agi.topicbox.com/groups/agi/
> I'll give this draft another read/bugfix before posting on Arxiv though...
yeah -- I'm gonna remove the qubit related stuff and put that in a
separate paper... I fixed some errors in that and did some more stuff
but it's getting too much for a side-note in this paper which is
focused on other ma
Yes an AGI* _in a box_* can attain more "data" by improving its intelligence.
Say we have a dataset it has in the box "The cat ate food. The dog ate food.
The cat meows." Now the AGI adds the ability to translate words cat=dog by
seeing they share some contexts (eat) so maybe they share others t
