Learning Cole-Hopf Coordinates for Stochastic Growth Dynamics: the KPZ class
Abstract
Inferring effective equations from observations of nonlinear stochastic processes is challenging because individual trajectories are noisy realizations, while the relevant physical information is encoded in ensemble statistics. We introduce a data-driven framework that learns a change of variables in which nonlinear stochastic dynamics become linear. The model combines a neural change of representation with a linear stochastic latent evolution and is trained through a short-time moment-matching objective, thereby avoiding trajectory \rev{simulation} during optimization. This construction yields an interpretable generative model: both the learned transformation and the latent evolution operators can be directly inspected. We validate the method on the one-dimensional Kardar--Parisi--Zhang (KPZ) equation, a paradigmatic model of stochastic growth for which the Cole--Hopf transformation maps the nonlinear dynamics to the stochastic heat equation. From simulated height trajectories alone, the model recovers the local exponential Cole--Hopf map and the associated linear stochastic operator. Generated trajectories reproduce the main statistical signatures of the KPZ universality class, including the growth law Var(ht)∼ t2β with β=1/3 and the GOE Tracy--Widom height distribution for flat initial conditions. These results show that interpretable stochastic equations can be inferred from data by learning representations in which the dynamics closes in a simple form. Furthermore the development of the physical informed learning permits to get insights on the physics of the dynamical process.
Related
- Fluctuations and multifractality in stochastic models of interface growth and population dynamics
- Learning a general class of admissible multi-species collision operators from molecular dynamics
- Speed Limit for Information Acquisition in Stochastic Learning Dynamics
- Beyond Synthesis Conditions: Linking In-situ Growth Dynamics to Carbon Nanotube Forest Properties through Multimodal Machine Learning
- Logit--Koopman coordinates for scale-resolved forecast-error growth in a three-level Lorenz-96 system
BibTeX
@misc{mensi2026learning,
title = {{Learning Cole-Hopf Coordinates for Stochastic Growth Dynamics: the KPZ class}},
author = {Lorenzo Mensi and Simone Cuonzo and Maria Chiara Angelini and Sergio Chibbaro and Cyril Furtlehner and Alberto Rosso},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.30919},
eprint = {2610.30919},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.30919},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Learning Cole-Hopf Coordinates for Stochastic Growth Dynamics: the KPZ class
AU - Lorenzo Mensi
AU - Simone Cuonzo
AU - Maria Chiara Angelini
AU - Sergio Chibbaro
AU - Cyril Furtlehner
AU - Alberto Rosso
PY - 2026
DA - 2026/10/22
PB - arXiv
DO - 10.48550/arxiv.2610.30919
UR - https://arxiv.org/abs/2610.30919
AB - Inferring effective equations from observations of nonlinear stochastic processes is challenging because individual trajectories are noisy realizations, while the relevant physical information is encoded in ensemble statistics. We introduce a data-driven framework that learns a change of variables in which nonlinear stochastic dynamics become linear. The model combines a neural change of representation with a linear stochastic latent evolution and is trained through a short-time moment-matching objective, thereby avoiding trajectory \rev{simulation} during optimization. This construction yields an interpretable generative model: both the learned transformation and the latent evolution operators can be directly inspected. We validate the method on the one-dimensional Kardar--Parisi--Zhang (KPZ) equation, a paradigmatic model of stochastic growth for which the Cole--Hopf transformation maps the nonlinear dynamics to the stochastic heat equation. From simulated height trajectories alone, the model recovers the local exponential Cole--Hopf map and the associated linear stochastic operator. Generated trajectories reproduce the main statistical signatures of the KPZ universality class, including the growth law Var(h^t)∼ t^2β with β=1/3 and the GOE Tracy--Widom height distribution for flat initial conditions. These results show that interpretable stochastic equations can be inferred from data by learning representations in which the dynamics closes in a simple form. Furthermore the development of the physical informed learning permits to get insights on the physics of the dynamical process.
ER -