Forecasting and predicting stochastic agent-based model data with biologically-informed neural networks

8 Nov 2023  ·  John T. Nardini ·

Stochastic agent-based models (ABMs) are widely used to describe collective processes in ecology, epidemiology, and cellular biology. It is challenging to predict how ABMs behave over many parameter values due to their computational nature. Modelers often address this limitation by coarse-graining ABM rules into mean-field differential equation (DE) models. These DE models are advantageous because they are fast to simulate; unfortunately, DE models can provide poor ABM predictions (or be ill-posed) in certain regions of parameter space. In this work, we describe how biologically-informed neural networks (BINNs) can be trained to learn BINN-guided PDE models to predict ABM behavior. In particular, we show that BINN-guided PDE simulations can forecast future ABM data not seen during model training, and we can predict ABM data at previously-unexplored parameter values by combining BINN-guided PDE simulations with multivariate interpolation. We demonstrate our approach using three case study ABMs of collective migration. We find through these case studies that BINN-guided PDEs accurately forecast and predict ABM data with a one-compartment PDE when the mean-field PDE is ill-posed or requires two compartments. This work is broadly applicable to studying biological systems that exhibit collective behavior. All code and data from our study is available at https://github.com/johnnardini/Forecasting_predicting_ABMs

PDF Abstract

Datasets


  Add Datasets introduced or used in this paper

Results from the Paper


  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.

Methods