Deterministic Global Optimization over Trained Kolmogorov-Arnold Networks (KANs)

​The repository “optimization-over-KANs” provides resources for deterministic global optimization over trained Kolmogorov-Arnold Networks (KANs). It includes Mixed-Integer Nonlinear Programming (MINLP) formulations implemented in Pyomo, Python scripts for training Multilayer Perceptrons (MLPs) using TensorFlow, and optimization routines for trained MLPs utilizing OMLT. The repository also offers the necessary code and data to reproduce the results presented in the associated research paper.

Open-source code

This is the source code you can use.