Resending: this message never made it to the list.
On 2019-02-04 19:32:PM, TimTyler wrote:
On 2019-02-03 22:07:PM, Matt Mahoney wrote:
Copying a bit requires deleting the old value. So Landauer's limit
applies.
That's not correct. Copying a bit doesn't require deleting the old bit.
Landauer's limit does not apply. Maybe you have some different notion
of "copying" from me.
As I use the term, copying involves information propagating to
multiple locations. If information
is more than one place - i.e. different locations share Shannon mutual
information - then
information was copied from one to the other - or from another source
to both places.
Reversible computing is free but really isn't useful for AGI. The
brain is not a quantum computer. Training a neural network performs
irreversible state changes.
Training a neural network does not require irreversible state changes
if you simulate it on a reversible computer.
That's possible since any Turing complete system can simulate
any other one, given enough space, time / memory.
--
__________
|im |yler http://timtyler.org/
------------------------------------------
Artificial General Intelligence List: AGI
Permalink:
https://agi.topicbox.com/groups/agi/Ta6fce6a7b640886a-Mfd0ec19a5d6fd0e56315049d
Delivery options: https://agi.topicbox.com/groups/agi/subscription