Variational Streaming Flow：在物理时间中进行概率预测

View PDF HTML (experimental)

Abstract:Probabilistic forecasting is important for predicting complex dynamical systems because intrinsic randomness and incomplete observations can cause the same observed state to evolve into multiple plausible futures. While flow matching is a flexible approach for probabilistic forecasting, it is computationally expensive. Streaming flow (SF) reformulates this approach to model temporal evolution efficiently by learning a continuous velocity field directly in physical time. However, SF learns a deterministic velocity field. Thus, it provides only a single future trajectory for a given fixed initial state and observation history. To overcome this limitation, we introduce Variational Streaming Flow (VSF). Our approach learns a latent distribution that is conditioned on the dynamics of interest. In turn, this enables probabilistic forecasting. Importantly, we retain the computational efficiency of SF by generating in physical time. Across deterministic and stochastic dynamical systems, VSF demonstrates superior predictive accuracy and distributional fidelity. We demonstrate the advantage for both long-horizon rollouts exceeding 1,000 steps, and settings with bifurcating dynamics. Moreover, VSF can be integrated into existing Joint-Embedding Predictive Architecture (JEPA)-based world models as a plug-and-play predictor to improve temporal dynamics and goal-directed success rate in navigation, motion planning, and manipulation.

Subjects:	Machine Learning (cs.LG)
Cite as:	arXiv:2610.00976 [cs.LG]
	(or arXiv:2610.00976v1 [cs.LG] for this version)
	https://doi.org/10.48550/arXiv.2610.00976 arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Hans Hao-Hsun Hsu [view email]
[v1] Thu, 1 Oct 2026 03:10:43 UTC (4,315 KB)