Continuous Normalizing Flow Training via Variational Vector Fields

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for training continuous normalizing flows (CNFs) are computationally costly and inefficient, particularly in scenarios where complex sensor data requires extensive preprocessing and lacks efficient anomaly detection capabilities.

Innovation Solution

Approximate the vector field of a CNF using variational inference in the flow matching method, enabling efficient training and improved anomaly detection by aligning generative paths with realistic trajectories.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional CNF training methods are used to solve ordinary differential equations, then accurate modeling of sensor data distributions is achieved, but computational cost and training time increase significantly

Engineering Contradiction:
Improvemodeling accuracyVSAvoidtraining time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent changes the training approach from solving ordinary differential equations to using flow matching with vector field approximation. This parameter change in the training methodology reduces computational complexity while maintaining modeling accuracy, directly resolving the contradiction between accurate modeling and training time

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent substitutes the traditional ODE-solving mechanism with a flow matching approach that uses vector field approximation. This replacement eliminates the need for computationally intensive numerical integration while achieving similar or better modeling results, addressing both accuracy and time efficiency

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Productivity

If flow matching method is used for CNF training, then training time is reduced to match supervised learning costs, but prediction capabilities and anomaly detection accuracy deteriorate

Engineering Contradiction:
Improvetraining speedVSAvoidprediction accuracy
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent introduces a variational distribution as an intermediary to approximate the true data distribution in flow matching. This intermediary enables the model to learn realistic trajectories and generative paths, improving prediction accuracy while maintaining the computational efficiency of flow matching

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent incorporates feedback mechanisms through the variational inference process, where the approximate posterior distribution is continuously refined during training. This feedback loop ensures that the vector field approximation aligns with feasible trajectories, enhancing both anomaly detection accuracy and generative capabilities

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4660888A1Device and method for training a continuous normalizing flow for processing sensor signals
Publication Date: 2025.12.10 ROBERT BOSCH GMBH
  • EP4660888A1 patent drawingFigure 1
  • EP4660888A1 patent drawingFigure 2
  • EP4660888A1 patent drawingFigure 3

AI summary

Computer-implemented method for training a continuous normalizing flow (60), wherein the continuous normalizing flow (60) is configured to a accept sensor signal (S) or a representation (30) of the sensor signal (S) as input and wherein the continuous normalizing flow (60) is trained using a flow matching method, characterized in that the vector field used in the flow matching method is approximated using variational inference.