Machine learning from limited data: Predicting biological dynamics under a time-varying external input

TL;DR

This paper demonstrates that reservoir computing can effectively predict complex, stochastic biological dynamics and infer unobserved conditions from limited data, including steady states and transient timescales.

Contribution

It introduces the application of reservoir computing to biological shape dynamics, showing its ability to predict steady states, transient timescales, and infer unobserved conditions from minimal data.

Findings

RC predicts steady state climate from limited data

RC learns transient timescales from four observations

RC infers statistics of unobserved cell shape conditions

Abstract

Reservoir computing (RC) is known as a powerful machine learning approach for learning complex dynamics from limited data. Here, we use RC to predict highly stochastic dynamics of cell shapes. We find that RC is able to predict the steady state climate from very limited data. Furthermore, the RC learns the timescale of transients from only four observations. We find that these capabilities of the RC to act as a dynamic twin allows us to also infer important statistics of cell shape dynamics of unobserved conditions.