DENG-MIT
Host public codes developed in Deng Energy and Nanotechnology Group | MIT
Built from DENG-MIT's public GitHub profile, Cambridge, MA. Not on Take My Repo yet, so there are no posts or updates here.
The code is associated with the paper entitled "KAN-ODEs: Kolmogorov-Arnold Network Ordinary Differential Equations for Learning Dynamical Systems and Hidden Physics"
Benchmark for learning stiff problems using physics-informed machine learning
A Parameter-Lean Kolmogorov-Arnold Network Layer with Improved Memory Efficiency and Convergence Behavior. See included link for full manuscript.
CoFFe stands for Collaboratively Fused Feature, a method designed for nonlinear data dimensionality reduction of multi-physics coupled systems.
Inference of chemical kinetics and thermodynamic properties from constant-volume combustion of energetic materials
Barebones ChemKAN implementation to accompany PCCP paper: "ChemKANs for combustion chemistry modeling and acceleration". Run in Julia 1.11.1.
Two stages pipeline for reduce combustion mechanisms: reduction and retrain