Simultaneous identification of models and parameters of scientific simulators
This provides a tool for scientists to make uncertainty-informed modeling decisions in compositional stochastic simulators, though it is incremental as it builds on simulation-based inference frameworks.
The paper tackles the problem of selecting model components in scientific simulators by introducing simulation-based model inference (SBMI), which trains neural networks to infer joint distributions over model components and parameters without likelihood evaluations. The result shows that SBMI can discover multiple data-consistent configurations and reveal non-identifiable components in neuroscience models.
Many scientific models are composed of multiple discrete components, and scientists often make heuristic decisions about which components to include. Bayesian inference provides a mathematical framework for systematically selecting model components, but defining prior distributions over model components and developing associated inference schemes has been challenging. We approach this problem in a simulation-based inference framework: We define model priors over candidate components and, from model simulations, train neural networks to infer joint probability distributions over both model components and associated parameters. Our method, simulation-based model inference (SBMI), represents distributions over model components as a conditional mixture of multivariate binary distributions in the Grassmann formalism. SBMI can be applied to any compositional stochastic simulator without requiring likelihood evaluations. We evaluate SBMI on a simple time series model and on two scientific models from neuroscience, and show that it can discover multiple data-consistent model configurations, and that it reveals non-identifiable model components and parameters. SBMI provides a powerful tool for data-driven scientific inquiry which will allow scientists to identify essential model components and make uncertainty-informed modelling decisions.