KPPax: preprocess differentiable kinetic ODE for machine learning in atmospheric chemistry
arXiv cs.AI · 28 Oct 2026, 03:58 AEST
KPPax is a new kinetic preprocessor that generates differentiable chemistry code in JAX from KPP language mechanisms. It enables parameter estimation and hybrid physics-machine-learning modeling in atmospheric chemistry.
Key facts
- Atmospheric chemical mechanisms are commonly distributed in the language of the Kinetic PreProcessor (KPP), which generates efficient Fortran or C code but does not fit into the differentiable programming frameworks in which machine-learning models are developed.
- KPPax evaluates the rate law with sparse operations in log space, which keeps it differentiable, and integrates the stiff system with the implicit solvers and the forward- and reverse-mode adjoints of the diffrax library.
- We demonstrate two applications with the POLLU benchmark mechanism.
- In the first, the Fisher information matrix computed from an automatically differentiated Jacobian predicts, before any fit, which of the 25 rate coefficients a flow-tube-like design can identify.
- In the second, a neural-network correction of an emission inventory is trained offline on per-scenario 4D-Var retrievals and then fine-tuned online through the chemical model.
Read aloud
KPPax is a new kinetic preprocessor that generates differentiable chemistry code in JAX from KPP language mechanisms. It enables parameter estimation and hybrid physics-machine-learning modeling in atmospheric chemistry. Atmospheric chemical mechanisms are commonly distributed in the language of the Kinetic PreProcessor KPP, which generates efficient Fortran or C code but does not fit into the differentiable programming frameworks in which machine-learning models are developed. KPPax evaluates the rate law with sparse operations in log space, which keeps it differentiable, and integrates the stiff system with the implicit solvers and the forward- and reverse-mode adjoints of the diffrax library. We demonstrate two applications with the POLLU benchmark mechanism. In the first, the Fisher information matrix computed from an automatically differentiated Jacobian predicts, before any fit, which of the twenty five rate coefficients a flow-tube-like design can identify. In the second, a neural-network correction of an emission inventory is trained offline on per-scenario 4D-Var retrievals and then fine-tuned online through the chemical model.
Sources in this cluster
- KPPax: preprocess differentiable kinetic ODE for machine learning in atmospheric chemistry
arXiv cs.AI · 28 Oct 2026, 03:58 AEST