“…These effects include multiscale behavior and anomalous behavior such as super-and sub-diffusion, making nonlocal models suitable for a broad class of engineering and scientific applications. Such applications include subsurface transport, 11,38,39,59,60 image processing, 1,33,42 multiscale and multiphysics systems, 6,8,27 magnetohydrodynamic, 58 phase transitions, 10,14,16 finance, 57,54 stochastic processes, 12,19,45,47,48 and, more recently, fractional back propagation in training neural networks, where a fractional gradient is used in place of the standard gradient. 69 Nonlocal models are characterized by integral operators acting on the values of a function on nonlocal neighborhoods; these are usually Euclidean balls of radius δ > 0, which is referred to as the horizon or interaction radius.…”