AI summaryⓘ
The authors developed Phys-JEPA, a method for predicting multiple related time-based variables in physical systems by combining deep learning and physics rules. Unlike previous methods that only apply physics constraints to the final output, Phys-JEPA enforces these rules inside the hidden parts of the model where the future state is represented. This approach helps the model make more physically meaningful predictions while still capturing complex dynamics. They tested Phys-JEPA on climate, traffic, and electricity data, showing it generally improves accuracy compared to standard methods. Their work suggests that adding physics knowledge directly to internal model states is helpful for understanding and forecasting temporal data.
Multivariate time-series forecastingDeep learningPhysics-informed modelsLatent state representationJoint-embedding predictive architectureMean squared errorTemporal dynamicsPhysical consistencyClimate dataTraffic forecasting
Authors
Weizhi Nie, Weichao Liu, Honglin Guo, Yuting Su
Abstract
Multivariate forecasting in physical systems requires models that predict coupled temporal variables while preserving meaningful state evolution. Deep forecasters can fit temporal correlations, and physics-informed models can regularize predictions with scientific constraints, but these directions are often connected only at the decoded-output level. As a result, the hidden predictive state that generates future trajectories may remain statistically useful but physically unstructured. We introduce Phys-JEPA, a physics-informed joint-embedding predictive architecture for multivariate time-series forecasting. Phys-JEPA learns a latent world model in which predictive states are decomposed into physical and residual components, and physical consistency is imposed directly on latent states and latent transitions rather than only on decoded forecasts. This formulation uses known physical variables to organize the representation space while retaining residual capacity for unresolved dynamics. On Jena Climate 2009--2016, Phys-JEPA reduces aggregate MSE from 0.12482 to 0.12273 and temperature MSE from 0.01892 to 0.01831 at H=24. On Traffic, full Phys-JEPA improves aggregate MSE over the supervised baseline across all tested horizons, reducing H=192 MSE from 0.800784 to 0.773873. On Electricity, the best variant depends on horizon: static latent consistency is strongest at H=24 and H=48, while full Phys-JEPA gives the best aggregate and target-variable MSE at H=192. These initial results suggest that moving physics-informed learning from output space to latent predictive state space is a promising direction for interpretable temporal world models.