My Math Forum  

Go Back   My Math Forum > College Math Forum > Calculus

Calculus Calculus Math Forum

LinkBack Thread Tools Display Modes
October 23rd, 2016, 12:27 PM   #1
Joined: Oct 2013

Posts: 20
Thanks: 0

How derive new class of polynomials?

We have $\displaystyle g_n(x_0)$ which is composition of functions $\displaystyle f_i(x_0,x_1,...x_{i-1}), i=1..n$ where $\displaystyle f_i(...)$ are basic arithmetic operations (add, sub, mul) on $\displaystyle x_k$ and constants. For example:
$\displaystyle x_1 = 2\cdot x_0*x_0 + x_0+0.567$
$\displaystyle x_2 = x_0x_1 + x_1 + 0.341$
$\displaystyle x_3 = 3.2x_2x_2+x_1+x_0$
this way is simply to reach polynomial degree of $\displaystyle 2^n$ using only n multiplications.
Where this can be used ? In approximation of non-rational functions like tangent, arcus tangent or cosinus.
Algorithm of Remez enable to find optimal polynomial degree n, but if is required big accuracy (>double) these polynomials have high degrees.
How is possible restrict all polynomials of degree 60 to polynomials derived by above formula with max 6 or 7 multiplications? and how to find optimal approximation polynomial only inside this narrow class of polynomials?
Generalization of this wil is adding operation "divide" and obtaining rational expressions.
Fast formula of sqrt just generates high degree rational expressions.
If would be possible find optimals rational expressions narorwed to n-div, k-mul, computing of non-rational functions like sinus or tangent would be fast as computing sqrt.
Borneq is offline  

  My Math Forum > College Math Forum > Calculus

class, derive, polynomials

Thread Tools
Display Modes

Similar Threads
Thread Thread Starter Forum Replies Last Post
Derive the Value of 'e' VisionaryLen Calculus 28 September 2nd, 2016 01:39 PM
Help me derive this. PlzzHelp Algebra 1 November 18th, 2015 10:08 AM
Derive marifb Calculus 1 June 12th, 2015 02:37 PM
Derive this dudetheman Calculus 2 January 18th, 2012 11:44 AM
Derive -x/y instereo911 Calculus 2 October 8th, 2009 02:55 PM

Copyright © 2019 My Math Forum. All rights reserved.