Physics-based machine learning for modeling stochastic IP3-dependent calcium dynamics
Machine Learning
2021-09-14 v1 Chemical Physics
Computational Physics
Neurons and Cognition
Abstract
We present a machine learning method for model reduction which incorporates domain-specific physics through candidate functions. Our method estimates an effective probability distribution and differential equation model from stochastic simulations of a reaction network. The close connection between reduced and fine scale descriptions allows approximations derived from the master equation to be introduced into the learning problem. This representation is shown to improve generalization and allows a large reduction in network size for a classic model of inositol trisphosphate (IP3) dependent calcium oscillations in non-excitable cells.
Cite
@article{arxiv.2109.05053,
title = {Physics-based machine learning for modeling stochastic IP3-dependent calcium dynamics},
author = {Oliver K. Ernst and Tom Bartol and Terrence Sejnowski and Eric Mjolsness},
journal= {arXiv preprint arXiv:2109.05053},
year = {2021}
}
Comments
26 pages