What does this research mean for the field?

A simultaneous approach formulating DAE-constrained parameter estimation as a fully discretized nonlinear programming problem, combined with a bi-level decomposition strategy, enables efficient and scalable training of neural differential-algebraic systems of equations. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To develop a simultaneous optimization approach for training neural differential-algebraic equations (DAEs) that avoids the stability and convergence limitations of traditional sequential methods.

August 19, 2026Computational Optimization and ApplicationsOpen Access

A simultaneous approach for training neural differential-algebraic systems of equations

View Full Paper

Ask AI

Bookmark

Authors

LLLaurens R. Lueg VAVictor Alves DSDaniel Schicksnus

Discussion

Member takes

Overview

Computational study demonstrates a simultaneous nonlinear programming framework for neural differential-algebraic systems, indicating improved stability and scalability across high-index models.

Key Points

To develop a simultaneous optimization approach for training neural differential-algebraic equations (DAEs) that avoids the stability and convergence limitations of traditional sequential methods.
Formulated the training problem as a fully discretized nonlinear program (NLP) that solves for neural network parameters and system trajectories simultaneously.
Developed a bi-level decomposition strategy that solves a discretized NLP at each iteration of an outer gradient descent loop to efficiently compute parameter sensitivities.
Enforced constraint satisfaction directly at discretization points, enabling successful training of higher-index DAE systems.
Demonstrated scalable parameter estimation on larger problem instances and larger dataset sizes via the bi-level optimization decomposition.

Cite This Study

Lueg et al. (2026) studied this question.

synapsesocial.com/papers/6a8562eb03308d306e2d5dfd https://doi.org/10.1007/s10589-026-00823-y

View Full Paper

Ask AI

Bookmark

August 19, 2026Computational Optimization and ApplicationsOpen Access

A simultaneous approach for training neural differential-algebraic systems of equations

View Full Paper

Ask AI

Bookmark

Authors

LLLaurens R. Lueg VAVictor Alves DSDaniel Schicksnus

Discussion

Member takes

Overview

Computational study demonstrates a simultaneous nonlinear programming framework for neural differential-algebraic systems, indicating improved stability and scalability across high-index models.

Key Points

To develop a simultaneous optimization approach for training neural differential-algebraic equations (DAEs) that avoids the stability and convergence limitations of traditional sequential methods.
Formulated the training problem as a fully discretized nonlinear program (NLP) that solves for neural network parameters and system trajectories simultaneously.
Developed a bi-level decomposition strategy that solves a discretized NLP at each iteration of an outer gradient descent loop to efficiently compute parameter sensitivities.
Enforced constraint satisfaction directly at discretization points, enabling successful training of higher-index DAE systems.
Demonstrated scalable parameter estimation on larger problem instances and larger dataset sizes via the bi-level optimization decomposition.

Cite This Study

Lueg et al. (2026) studied this question.

synapsesocial.com/papers/6a8562eb03308d306e2d5dfd https://doi.org/10.1007/s10589-026-00823-y

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark