aiqbench analysis
This research introduces a novel surrogate modeling framework that accelerates dynamic simulations of stochastic systems by combining variational autoencoders with temporal fusion transformers. By effectively reducing dimensionality and capturing both short- and long-range dependencies, the method achieves simulation accuracy at a fraction of the computational cost while providing built-in uncertainty quantification. The approach addresses critical bottlenecks in high-dimensional feature spaces where traditional time series models fail.
Accelerated surrogate dynamics for dynamical, stochastic system evolution
arXiv:2609.37184v1 Announce Type: new Abstract: Dynamic simulations are an entrenched way of gaining insight into the evolution of system dynamics. Their computational cost however is often prohibitively high, especially in cases of stochastic frameworks. Machine learning algorithms are especially suited as simulation surrogates. Nevertheless, they face some very distinct limitations. Firstly, the sheer dimensionality of these systems, however, precludes the use of traditional time series models who struggle with high dimensional feature spaces. Additionally, traditional time series focus exclusively on either long or short range effects, causing local or global drift given enough time. In this paper, we propose a framework that addresses those limitations. Our framework combines a Variational Autoencoder, with a convolutional or graph basis that reduces the dimensionality of the system. This latent vector is propagated in time using a Temporal Fusion Transformer model, which includes both long range and short range effect encoding, as well as static covariate support. We test our framework on three distinct cases, to prove its robustness and in all three we have achieved practically identical to the simulation results at a fraction of the time. Further, our framework is flexible enough to be adapted to any new system and provides an inbuilt uncertainty quantification for targeted experiment design.
Abstract:Dynamic simulations are an entrenched way of gaining insight into the evolution of system dynamics. Their computational cost however is often prohibitively high, especially in cases of stochastic frameworks. Machine learning algorithms are especially suited as simulation surrogates. Nevertheless, they face some very distinct limitations. Firstly, the sheer dimensionality of these systems, however, precludes the use of traditional time series models who struggle with high dimensional feature spaces. Additionally, traditional time series focus exclusively on either long or short range effects, causing local or global drift given enough time. In this paper, we propose a framework that addresses those limitations. Our framework combines a Variational Autoencoder, with a convolutional or graph basis that reduces the dimensionality of the system. This latent vector is propagated in time using a Temporal Fusion Transformer model, which includes both long range and short range effect encoding, as well as static covariate support. We test our framework on three distinct cases, to prove its robustness and in all three we have achieved practically identical to the simulation results at a fraction of the time. Further, our framework is flexible enough to be adapted to any new system and provides an inbuilt uncertainty quantification for targeted experiment design.
Artificial Intelligence (cs.AI)
arXiv:2609.37184 [cs.AI]
(or arXiv:2609.37184v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.37184 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 296 words
Related coverage
Physics-informed neural operator for parametric phase-field modelling of interfacial degradation and microstructural evolution
arXiv · Oct 6, 2026
Reducing Spatial and Temporal Dimensionality in the Multidimensional Caldeira-Leggett Model
Quantum Journal · Sep 2, 2026
Inverse Design of Quantum Control Sequences with Fourier Neural Operators
arXiv · Aug 5, 2026
Information-Proliferating Dynamics
LessWrong · Oct 9, 2026