From: Nick Tarleton (nickptar@gmail.com)
Date: Tue Nov 25 2008 - 18:12:58 MST
On Tue, Nov 25, 2008 at 4:10 PM, Matt Mahoney <matmahoney@yahoo.com> wrote:
> --- On Tue, 11/25/08, Peter de Blanc <peter@spaceandgames.com> wrote:
>
> > Matt, your paper ends in the conclusion, "it seems
> > that even though RSI is possible in a mathematical sense, it
> > is too slow to be of practical use, or to compete with
> > evolutionary systems, at least for fixed goals." What
> > is the justification for this statement?
>
> The rate of information gain for RSI is O(log t).
Why do you identify intelligence with algorithmic complexity, again?
http://www.overcomingbias.com/2008/11/complexity-and.html
> The information gain for evolution is O(t), or 1 bit per population
> doubling and selection.
Really? Evolution isn't magic, and the suggestion that it can provably beat
all forms of intelligence is extremely suspicious. I can run an evolving
population in a closed simulation, where all the bits not describing the
evolving agents and their environment start as 0, or something similarly
low-complexity, and then apply your proof. This is for the whole population,
and picking out a random individual would require individual information
growing at O(a^t), but picking out the maximal individual(s) according to
some utility function would only add a constant factor (complexity of the
function).
npt
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:01:03 MDT