Flow Matching for Scalable Simulation-Based Inference

Oct-8-2025, 10:47:07 GMT–Neural Information Processing Systems

Figure 1: Comparison of network architectures (left) and flow trajectories (right). Discrete flows (NPE, top) require a specialized architecture for the density estimator. Continuous flows (FMPE, bottom) are based on a vector field parametrized with an unconstrained architecture.

architecture, arxiv, inference, (13 more...)

Neural Information Processing Systems

Oct-8-2025, 10:47:07 GMT

Conferences PDF

Add feedback

Country:
- Asia
  - Japan (0.04)
  - Taiwan (0.04)
- Europe
  - Hungary (0.04)
  - Portugal (0.04)
  - Ireland (0.04)
  - Belgium (0.04)
  - Greece (0.04)
  - United Kingdom > England
    - Nottinghamshire > Nottingham (0.14)
  - Monaco (0.04)
  - Spain (0.04)
  - Germany
    - Baden-Württemberg > Tübingen Region
      - Tübingen (0.14)
    - Bavaria > Upper Bavaria
      - Munich (0.04)
    - Lower Saxony (0.04)
  - Poland (0.04)
- North America
  - Canada > Quebec
    - Montreal (0.04)
  - United States (0.28)

Genre:
- Research Report > New Finding (0.46)

Technology:
- Information Technology
  - Artificial Intelligence
    - Machine Learning
      - Learning Graphical Models > Directed Networks
        Bayesian Learning (0.68)
      - Neural Networks (0.93)
      - Statistical Learning (1.00)
    - Representation & Reasoning > Uncertainty
      - Bayesian Inference (0.94)
  - Communications > Networks (0.66)
  - Data Science (0.93)

Duplicate Docs Excel Report

Title
Flow Matching for Scalable Simulation-Based Inference Jonas Wildberger

Similar Docs Excel Report more

Title	Similarity	Source
None found