Multiscale functions, Scale dynamics and Applications to partial differential equations
Jacky Cresson, Fr\'ed\'eric Pierret

TL;DR
This paper introduces multiscale functions, scale calculus, and scale dynamics based on time-scale calculus to understand how different continuous models emerge from data at various scales, especially in PDEs.
Contribution
It develops a formalism connecting scale regimes with different PDE models, including deriving classical and non-linear equations from scale Newton's equations.
Findings
Derived non-linear diffusion and Schrödinger equations from scale Newton's equations.
Explained the existence of multiple models for scale-invariant equations.
Revealed how classical equations are recovered under specific scale assumptions.
Abstract
Modeling phenomena from experimental data, always begin with a \emph{choice of hypothesis} on the observed dynamics such as \emph{determinism}, \emph{randomness}, \emph{derivability} etc. Depending on these choices, different behaviors can be observed. The natural question associated to the modeling problem is the following : \emph{"With a finite set of data concerning a phenomenon, can we recover its underlying nature ?} From this problem, we introduce in this paper the definition of \emph{multi-scale functions}, \emph{scale calculus} and \emph{scale dynamics} based on the \emph{time-scale calculus} (see \cite{bohn}). These definitions will be illustrated on the \emph{multi-scale Okamoto's functions}. The introduced formalism explains why there exists different continuous models associated to an equation with different \emph{scale regimes} whereas the equation is \emph{scale…
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.
