{"id":"arxiv/2606.28587","url":"https://arxiv.org/abs/2606.28587","tags":["math.NA"],"score":0,"title":"Constrained Kolmogorov widths","domain":"math.NA","source":"arxiv","authors":["Ronald DeVore","Guergana Petrova","Jonathan W. Siegel","Przemysław Wojtaszczyk"],"reviews":0,"subject":"math.NA","summary":"The main theme of approximation theory is to understand how well a general function $f$ can be approximated by a simpler function $g$ such as a polynomial or spline. In many applications, one wants $g$ to retain known properties of $f$ such as its inherent smoothness or a geometrical property such as monotonicity or convexity. Additional requirements on $g$ of this type are known as constraints. In this paper, we do a systematic study of constrained approximation to understand how the imposition of such constraints limits the efficiency of the approximation. We study constrained…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-06-26","reactions":{}}