Re: Simplification: science or heuristics?
From: Radu Zapotinschi (raduz_at_math.ubbcluj.ro)
Date: 01/20/05
- Next message: parisse_at_domain.invalid: "fast multivariate GCD algorithm (was Steps towards writing a computer algebra system /bad ideas from books)"
- Previous message: carlos_at_colorado.edu: "Re: Simplification: science or heuristics?"
- In reply to: Richard Fateman: "Re: Simplification: science or heuristics?"
- Next in thread: gmatthew_at_amgen.com: "Re: Simplification: science or heuristics?"
- Messages sorted by: [ date ] [ thread ]
Date: Thu, 20 Jan 2005 19:46:22 +0000 (UTC)
Science or heuristics? Both.
1) simplification is in general undecidable
(see http://mathworld.wolfram.com/RichardsonsTheorem.html
and
Caviness "On Canonical Forms and Simplification."
J. Assoc. Comp. Mach. 17, 385-396, 1970.)
2) simplification is context dependent, i.e., what is the simplest
form of an expression depends on what are you using the expression
for.
If you want to try a top-down approach to the bibliography, see:
http://www.cas.mcmaster.ca/~carette/publications/simplification.pdf
In practice, the standard simplification procedure of you CAS should
be the last resort. You should try other rewrite functions first,
or build your own combination of transformations starting from:
- the form of the expressions you get
- the form of the expressions you expect
That's easy for me to say :)
If you can be more specific on the problem you have and the CAS
you're using, then maybe someone out here will be able to give you a
more specific advice.
- Next message: parisse_at_domain.invalid: "fast multivariate GCD algorithm (was Steps towards writing a computer algebra system /bad ideas from books)"
- Previous message: carlos_at_colorado.edu: "Re: Simplification: science or heuristics?"
- In reply to: Richard Fateman: "Re: Simplification: science or heuristics?"
- Next in thread: gmatthew_at_amgen.com: "Re: Simplification: science or heuristics?"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
