The higher order partial derivatives of Okamoto's function with respect to the parameter
Pieter Allaart, Nathan Dalaklis, Kiko Kawamura, Matthew Ortiz, Jiajie Zheng

TL;DR
This paper studies the higher order partial derivatives of Okamoto's family of self-affine functions, analyzing their fractal dimensions, differentiability properties, and points of infinite derivative, revealing new phenomena beyond known properties.
Contribution
It introduces and analyzes the higher order derivatives of Okamoto's functions, providing new insights into their fractal and differentiability characteristics.
Findings
Computed the box-counting dimension of the derivatives' graphs
Characterized the differentiability of the derivatives
Identified points where derivatives are infinite
Abstract
Let be Okamoto's family of continuous self-affine functions, introduced in [{\em Proc. Japan Acad. Ser. A Math. Sci.} {\bf 81} (2005), no. 3, 47--50]. This family includes well-known ``pathological" examples such as Cantor's devil's staircase and Perkins' continuous but nowhere differentiable function. It is well known that is real analytic in for every . We introduce the functions \[ M_{k,a}(x):=\frac{\partial^k}{\partial a^k}F_a(x), \qquad k\in\mathbb{N}, \quad x\in[0,1]. \] We compute the box-counting dimension of the graph of , characterize its differentiability, and investigate in detail the set of points where has an infinite derivative. While some of our results are similar to the known facts about Okamoto's function, there are also some notable differences and surprising new phenomena that arise when considering the…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
