Learning probabilistic models of biological systems us-ing active inference with belief propagation

ERNESTO MARTÍNEZ

Buenos Aires

Simposio; SIMPOSIO ARGENTINO DE CIENCIA DE DATOS Y GRANDES DATOS (JAIIO); 2021

SADIO

In this work, the normative framework of active inference is integrat-ed with belief propagation for inverting a probabilistic causal model using data generated from planned interactions between a Bayesian modeling agent and a biological system. Thompson sampling of parameter distributions is used to es-timate the free energy of the expected future when beliefs about beliefs are rolled over a planning horizon. Learning a probabilistic model for maximizing biomass production in the well-known Baker?s yeast example is used as an ex-ample. The prior parameter distributions in the system model of a fed-batch cul-tivation are updated as new observations are obtained. Planned action sequenc-es aim to excite the yeast metabolism by introducing changes in the feed rate of two nutrients (glucose and nitrogen). Results obtained demonstrate that by max-imizing the model evidence, the proposed approach constraints biological sys-tem dynamics to relevant trajectories for improved parametric precision in the preferred region of physiological states that favor biomass productivity.