RSS 2.0 Feed

» Welcome Guest Log In :: Register

Pages: (501) < [1] 2 3 4 5 6 ... >   
  Topic: Uncommonly Dense Thread 3, The Beast Marches On...< Next Oldest | Next Newest >  
Zachriel



Posts: 2723
Joined: Sep. 2006

(Permalink) Posted: July 27 2010,07:13   

Quote
DATCG: Dissecting AVIDA simulation of evolution

The paper dissects AVIDA and finds Active Information that guides the performance of “simulated evolution.”

Quote
Ewert, Dembski & Marks: Even moderately sized search problems require external assistance to be successful. The contribution of the external source to the search can be measured as active information. Sources of active information include (a) an oracle that provides fitness information (or scores) for the search, ...

As has been pointed out many times before, the oracle represents the abstracted environment, and is the whole point of many evolutionary algorithms.

Quote
Ewert, Dembski & Marks: Since every constructive choice of priors has a corresponding deleterious choice, the average performance over all searches is the same as a random search [3], [4], [5], [8], [15], [19], [20], [21], [22], [23], [27], [28], [29], [30]. This search property is dubbed conservation of information [4], [5], [7], [9], [23] as popularized by the no free lunch theorem [3], [8], [15], [30].

This is highly misleading with respect to evolution. Adding zillions of footnotes doesn't resolve the ambiguity. We're not concerned with a highly chaotic search space, but one that is highly ordered. Nor are we discussing an arbitrary algorithm, but one adept at climbing fitness gradients.

--------------

You never step on the same tard twice—for it's not the same tard and you're not the same person.

   
  15001 replies since Sep. 04 2009,16:20 < Next Oldest | Next Newest >  

Pages: (501) < [1] 2 3 4 5 6 ... >   


Track this topic Email this topic Print this topic

[ Read the Board Rules ] | [Useful Links] | [Evolving Designs]