Abstract:Forecasting stochastic nonlinear dynamical systems under the influence of conditioning variables is a fundamental challenge repeatedly encountered across the biological and physical sciences. While flow-based models can impressively predict the temporal evolution of probability distributions representing possible outcomes of a specific process, existing frameworks cannot satisfactorily account for the impact of conditioning variables on these dynamics. Amongst several limitations, existing methods require training data with paired conditions and are developed for discrete conditioning variables. We propose Conditional Variable Flow Matching (CVFM), a framework for learning flows transforming conditional distributions with amortization across continuous conditioning variables - permitting predictions across the conditional density manifold. This is accomplished through several novel advances, in particular, simultaneous sample conditioned flows over the main and conditioning variables, alongside a conditional Wasserstein distance and kernel facilitating conditional optimal transport. Collectively, these advances allow for learning system dynamics provided measurement data whose states and conditioning variables are not in correspondence. We demonstrate CVFM on a suite of increasingly challenging problems, including discrete and continuous conditional mapping benchmarks, image-to-image domain transfer, and modeling the temporal evolution of materials internal structure during manufacturing processes. We observe that CVFM results in improved performance and convergence characteristics over alternative conditional variants.

Flow Map Learning for Unknown Dynamical Systems: Overview, Implementation, and Benchmarks

Principal Component Flow Map Learning of PDEs from Incomplete, Limited, and Noisy Data

Flow Map Matching

ImitationFlow: Learning Deep Stable Stochastic Dynamic Systems by Normalizing Flows

Integration-free Learning of Flow Maps

D^3FlowSLAM: Self-Supervised Dynamic SLAM with Flow Motion Decomposition and DINO Guidance

Low-dimensional Flow Models from high-dimensional Flow data with Machine Learning and First Principles

Conditional Variable Flow Matching: Transforming Conditional Densities with Amortized Conditional Optimal Transport

Recent Progress of Machine Learning in Flow Modeling and Active Flow Control

Switched Flow Matching: Eliminating Singularities via Switching ODEs

Flow Matching for Scalable Simulation-Based Inference

Unifying Predictions of Deterministic and Stochastic Physics in Mesh-reduced Space with Sequential Flow Generative Model

Explore missing flow dynamics by physics-informed deep learning: The parameterized governing systems

Model-based deep reinforcement learning for accelerated learning from flow simulations

Bayesian Flow Networks in Continual Learning

Explore missing flow dynamics by physics-informed deep learning: the parameterised governing systems

Learning dynamical systems from data: A simple cross-validation perspective, Part V: Sparse Kernel Flows for 132 chaotic dynamical systems

Generalized Flow Matching for Transition Dynamics Modeling

Unifying Bayesian Flow Networks and Diffusion Models through Stochastic Differential Equations

Flow Matching for Generative Modeling