Re: encoding problem
- From: Patrick May <pjm@xxxxxxx>
- Date: Sun, 03 Sep 2006 18:27:00 +0100
Patrick May <pjm@xxxxxxx> writes:
The only other obvious approach I can suggest without knowing
more details of the domain is to include in the chromosome a few bits
for each of the variables and only treat the four with the highest
values as active.
Rudely following up to myself, another option occurred to me just
as I sent the last post. The number of unique combinations of four
elements taken from ten is (10! / 4! * (10 - 4)!), or 210. This fits
comfortably into eight bits. If you map each number in the range
0-209 to a specific combination and include it in your chromosome, you
should be able to evolve only allowed solutions.
Regards,
Patrick
------------------------------------------------------------------------
S P Engineering, Inc. | Large scale, mission-critical, distributed OO
| systems design and implementation.
pjm@xxxxxxx | (C++, Java, Common Lisp, Jini, middleware, SOA)
.
- Follow-Ups:
- Re: encoding problem
- From: Oliver Gobin
- Re: encoding problem
- References:
- encoding problem
- From: Oliver Gobin
- Re: encoding problem
- From: Patrick May
- Re: encoding problem
- From: Oliver Gobin
- Re: encoding problem
- From: Patrick May
- encoding problem
- Prev by Date: Re: encoding problem
- Next by Date: Re: encoding problem
- Previous by thread: Re: encoding problem
- Next by thread: Re: encoding problem
- Index(es):