Deep Learning for Fast Inference of Mechanistic Models' Parameters

TL;DR

This paper introduces a deep learning approach to rapidly estimate parameters of mechanistic models in bioprocess engineering, significantly reducing computation time while maintaining or improving accuracy over traditional fitting methods.

Contribution

The authors propose a neural network-based method for direct parameter prediction from observations, combining mechanistic models with deep learning to enhance efficiency and accuracy.

Findings

Neural network estimates are slightly improved by further fitting.

The proposed method outperforms traditional gradient-based fitting in speed.

Once trained, the neural network provides parameter estimates orders of magnitude faster.

Abstract

Inferring parameters of macro-kinetic growth models, typically represented by Ordinary Differential Equations (ODE), from the experimental data is a crucial step in bioprocess engineering. Conventionally, estimates of the parameters are obtained by fitting the mechanistic model to observations. Fitting, however, requires a significant computational power. Specifically, during the development of new bioprocesses that use previously unknown organisms or strains, efficient, robust, and computationally cheap methods for parameter estimation are of great value. In this work, we propose using Deep Neural Networks (NN) for directly predicting parameters of mechanistic models given observations. The approach requires spending computational resources for training a NN, nonetheless, once trained, such a network can provide parameter estimates orders of magnitude faster than conventional methods. We consider a training procedure that combines Neural Networks and mechanistic models. We demonstrate the performance of the proposed algorithms on data sampled from several mechanistic models used in bioengineering describing a typical industrial batch process and compare the proposed method, a typical gradient-based fitting procedure, and the combination of the two. We find that, while Neural Network estimates are slightly improved by further fitting, these estimates are measurably better than the fitting procedure alone.