.png){kind=icon}
.png){kind=icon}
Return
Partially-supervised neural network model for quadratic multiparametric programming
.png){kind=icon}
OA
AI
DOI：10.1016/j.cor.2026.107564.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• We propose a NN model that exactly represents the solution function of QP. • We explain inaccuracies of traditional black-box NN-based QP solvers. • We derive some model parameters directly from the problem coefficients. • Proof of concept support for the method is provided using empirical datasets. • The model produces solutions instantly with less than 1E-10 squared KKT violations.
Keywords:
Partially-supervised neural network
Analytical derivation of model parameters
Multiparametric programming
Neural network for optimization
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
C
IF:
4.3
Papers:
201
Citations:
0
Organization
No organization information available