Learning chemical reaction networks from trajectory data

13 Feb 2019 · Wei Zhang, Stefan Klus, Tim Conrad, Christof Schütte ·

We develop a data-driven method to learn chemical reaction networks from trajectory data. Modeling the reaction system as a continuous-time Markov chain and assuming the system is fully observed, our method learns the propensity functions of the system with predetermined basis functions by maximizing the likelihood function of the trajectory data under $l^1$ sparse regularization. We demonstrate our method with numerical examples using synthetic data and carry out an asymptotic analysis of the proposed learning procedure in the infinite-data limit.

PDF Abstract

Code

Add Remove Mark official

zwpku/sparse-learning-CRN official

Edit Social Preview

Learning chemical reaction networks from trajectory data

Code Edit Add Remove Mark official

Categories

Code

Add Remove Mark official