From: Eliezer S. Yudkowsky (sentience@pobox.com)
Date: Tue Mar 13 2001 - 11:38:31 MST
Dale Johnstone wrote:
>
> BTW don't forget, this is in the context of AI - linear improvements
> can lead to exponential returns. They don't call it a Singularity for
> nothing! :)
Yeah, but *that's* not automatic. It only works if you *already* *have* a
self-modifying AI - and one that's not in the middle of a bottleneck;
otherwise it just does nothing faster. I suppose that compressing the RAM
might be OK if it were totally transparent to the programmer, and I'm sure
that seed AIs past the self-optimizing compiler stage will think
pre-compressed thoughts, but aside from that, I think I'm basically in
agreement with James Rogers; performance tuning is an addictive,
distracting, and self-indulgent habit (yes, I had it too).
-- -- -- -- --
Eliezer S. Yudkowsky http://intelligence.org/
Research Fellow, Singularity Institute for Artificial Intelligence
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:36 MDT