Functional-Hybrid modeling through automated adaptive symbolic regression for interpretable mathematical expressions

Abstract

Mathematical models used for the representation of (bio)-chemical processes can be grouped into two broad paradigms: white-box or mechanistic models, completely based on knowledge or black-box data-driven models based on patterns observed in data. However, in the past twodecade, hybrid modeling that explores the synergy between the two paradigms has emerged as a pragmatic compromise. The data-driven part of these have been largely based on conventional machine learning algorithm (e.g., artificial neural network, support vector regression), which prevents interpretability of the finally learnt model by the domain-experts.In this work we present a novel hybrid modeling framework, the Functional-Hybrid model, that uses the ranked domain-specific functional beliefs together with symbolic regression to develop dynamic models. We demonstrate the successful implementation of these hybrid models for four benchmark systems and a microbial fermentation reactor, all of which are systems of (bio)chemical relevance. We also demonstrate that compared to a similar implementation with the conventional ANN, the performance of Functional-Hybrid model is at least two times better in interpolation and extrapolation. Additionally, the proposed framework can learn the dynamics in 50% lower number of experiments. This improved performance can be attributed to the structure imposed by the functional transformations introduced in the Functional-Hybrid model.

A Symbolic Regression Approach to Hepatocellular Carcinoma Diagnosis Using Hypermethylated CpG Islands in Circulating Cell-Free DNA

“…Compared to simple biomarker analysis, machine learning techniques display better performance, which is expected due to their more complex and opaque nature. The symbolic regression approach described in this paper, however, attains similar sensitivity and specificity to conventional ML approaches, is much more transparent and allows for mathematical reasoning of the results in a biological context, which can be a source of useful further research (Narayanan, Cruz Bournazou, Guillén Gosálbez, & Butté, 2022; Cardoso et al, 2020).…”

Hybrid Models Based on Machine Learning and an Increasing Degree of Process Knowledge: Application to Cell Culture Processes

“…A further detailed description of the training procedure for developing the hybrid models can be found in our previous work. , …”

“…Successful application cases include various chemical processes, − energy storage and management, and several areas of biotechnology ,,,− including fermentations and cell cultures. ,, Their use has been emphasized and demonstrated for different process applications such as optimization, − quality modeling, monitoring, − and control. ,,, Subsequently, it is considered as a key future technology for the realization of Quality by DesignProcess Analytical Technology (QbD-PAT) and industry 4.0 − initiatives.…”

Machine learning interatomic potential: Bridge the gap between small-scale models and realistic device-scale simulations

“… 134 Additional conformations are collected when model predictions become unreliable, based on an uncertainty criterion, and reference calculations are performed. 135 , 136 Meta dynamics Sampling: similar to adaptive sampling, this method uses preliminary ML potentials in MD simulations but biases the dynamics to visit unexplored regions on the PES. 137 , 138 , 139 It combines metadynamics with uncertainty estimates to select relevant structures.…”