A Metaheuristic for Amortized Search in High-Dimensional Parameter Spaces
This work addresses a challenging problem in (bio)physical systems modeling, offering an incremental improvement for researchers dealing with high-dimensional parameter spaces.
The paper tackles the problem of parameter inference in high-dimensional, non-linear dynamical models by proposing a metaheuristic called DR-FFIT that uses feature-informed transformations for dimensionality reduction, resulting in improved performance over existing metaheuristics and better model fit with contained run-time costs.
Parameter inference for dynamical models of (bio)physical systems remains a challenging problem. Intractable gradients, high-dimensional spaces, and non-linear model functions are typically problematic without large computational budgets. A recent body of work in that area has focused on Bayesian inference methods, which consider parameters under their statistical distributions and therefore, do not derive point estimates of optimal parameter values. Here we propose a new metaheuristic that drives dimensionality reductions from feature-informed transformations (DR-FFIT) to address these bottlenecks. DR-FFIT implements an efficient sampling strategy that facilitates a gradient-free parameter search in high-dimensional spaces. We use artificial neural networks to obtain differentiable proxies for the model's features of interest. The resulting gradients enable the estimation of a local active subspace of the model within a defined sampling region. This approach enables efficient dimensionality reductions of highly non-linear search spaces at a low computational cost. Our test data show that DR-FFIT boosts the performances of random-search and simulated-annealing against well-established metaheuristics, and improves the goodness-of-fit of the model, all within contained run-time costs.