Sample efficient likelihood-free inference for virus dynamics with different types of experiments

TL;DR

This paper introduces an improved likelihood-free inference method using Bayesian optimization for virus dynamics, enabling efficient parameter estimation with uncertainty quantification across diverse experimental data types.

Contribution

It presents an enhanced BOLFI algorithm with Gaussian process classifiers and a discrepancy design for integrating heterogeneous data, applied successfully to influenza A virus data.

Findings

Remarkable computational efficiency over existing methods

Effective parameter estimation for influenza A virus

Successful integration of diverse experimental data

Abstract

This study applied Bayesian optimization likelihood-free inference(BOLFI) to virus dynamics experimental data and efficiently inferred the model parameters with uncertainty measure. The computational benefit is remarkable compared to existing methodology on the same problem. No likelihood knowledge is needed in the inference. Improvement of the BOLFI algorithm with Gaussian process based classifier for treatment of extreme values are provided. Discrepancy design for combining different forms of data from completely different experiment processes are suggested and tested with synthetic data, then applied to real data. Reasonable parameter values are estimated for influenza A virus data.