A Renormalization-Group Hierarchy of Stochastic Effective Dynamics Learned through Path Integrals
Abstract
This study combines a stochastic renormalization group (RG) in space with a path-integral description of the time evolution, giving a spatial hierarchy of coarse-grained dynamics together with the distribution over spatiotemporal paths. The RG is defined as a diffusion process that applies scale-dependent Laplacian damping together with additive Gaussian noise. The time evolution at each spatial scale is formulated as an Onsager--Machlup action, with a drift (i.e., the predictor) and white noise whose amplitude is fixed by the RG. The predictor of these dynamics is optimized by minimizing the Kullback--Leibler divergence between the path distributions from the RG and from the path-integral description. The optimal predictor contains the score function that connects the spatial scales, so the predictor and the score are two aspects of the same multiscale path formulation. The formulation unifies simulation using the predictor, unconditional generation using the score, and super-resolution using both. The predictor has no closed form, so it is computed by a neural network in two realizations that differ only in how the score is computed. The first realization obtains the score by automatic differentiation of the path distribution, remaining faithful to that formulation. The second obtains the score as an additional network output trained by denoising, at a lower computational cost. These two realizations are validated through numerical experiments on two representative multiscale systems, the Kolmogorov flow and the two-timescale Lorenz-96 model.
Related
- Regularization of stochastic fluid dynamics and non-perturbative renormalization of the shear viscosity of incompressible fluids
- The Renormalization Group as a Stochastic Exploration Process
- Scaling and Renormalization in Neural Networks: An Effective Renormalization-Group Description of Grokking
- Electroweak Thermal Histories in the Minimal Type-II Seesaw Model with a Renormalization Group-Improved Effective Potential
- The renormalization of the shell-model neutrinoless double-beta decay operator starting from effective field theory at leading order
BibTeX
@misc{yasuda2026renormalizationgroup,
title = {{A Renormalization-Group Hierarchy of Stochastic Effective Dynamics Learned through Path Integrals}},
author = {Yuki Yasuda and Tobias Bischoff},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.06099},
eprint = {2610.06099},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.06099},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - A Renormalization-Group Hierarchy of Stochastic Effective Dynamics Learned through Path Integrals
AU - Yuki Yasuda
AU - Tobias Bischoff
PY - 2026
DA - 2026/10/05
PB - arXiv
DO - 10.48550/arxiv.2610.06099
UR - https://arxiv.org/abs/2610.06099
AB - This study combines a stochastic renormalization group (RG) in space with a path-integral description of the time evolution, giving a spatial hierarchy of coarse-grained dynamics together with the distribution over spatiotemporal paths. The RG is defined as a diffusion process that applies scale-dependent Laplacian damping together with additive Gaussian noise. The time evolution at each spatial scale is formulated as an Onsager--Machlup action, with a drift (i.e., the predictor) and white noise whose amplitude is fixed by the RG. The predictor of these dynamics is optimized by minimizing the Kullback--Leibler divergence between the path distributions from the RG and from the path-integral description. The optimal predictor contains the score function that connects the spatial scales, so the predictor and the score are two aspects of the same multiscale path formulation. The formulation unifies simulation using the predictor, unconditional generation using the score, and super-resolution using both. The predictor has no closed form, so it is computed by a neural network in two realizations that differ only in how the score is computed. The first realization obtains the score by automatic differentiation of the path distribution, remaining faithful to that formulation. The second obtains the score as an additional network output trained by denoising, at a lower computational cost. These two realizations are validated through numerical experiments on two representative multiscale systems, the Kolmogorov flow and the two-timescale Lorenz-96 model.
ER -