This is the talk page for discussing improvements to the Iteratively re-weighted least squares article.