{"id":"arxiv/2608.25084","url":"https://arxiv.org/abs/2608.25084","tags":["cs.LG","math.NA"],"score":0,"title":"The Frame Kernel Method for Multiscale Operator Learning","domain":"cs.LG","source":"arxiv","authors":["Branden Frieden","Ryan Whitehead","M. Keith Ballard","Robert M. Kirby","Varun Shankar"],"reviews":0,"subject":"cs.LG","summary":"We present a natively multiscale operator learning method for the surrogate modeling of (numerical solvers for) multiscale partial differential equations (PDEs). The primary novelty of our method lies in a novel multiscale kernel frame function approximation technique. Leveraging this new kernel frame technique, we cast the operator learning problem as one of learning frame coefficients of output functions as a function of frame coefficients of input functions. The generalization step then automatically allows for a multiscale decomposition of the output functions. Our method is applicable…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-08-25","reactions":{}}