Re: What has become of John Maddock's implementation of the Remez algorithm for Chebychev approximation?

From: Bjorn Reese (breese_at_[hidden])
Date: 2021-06-18 14:47:03

On 6/16/21 3:03 PM, Joachim Wuttke via Boost wrote:

> If I understand correctly, function 
> chebyshev_transform::chebyshev_transform
> solves the minimax approximation problem in a completely different way than
> the Remez algorithm, but results should be of similar quality?

If I remember Trefetchen correctly, Remez reduces the worst-case error
(infinity norm) and Chebyshev reduces the average error (L_2 norm.)