Investigating the Impact of Model Misspecification in Neural Simulation-based Inference
This addresses a critical gap for researchers using neural SBI in scientific applications, highlighting a major limitation that could lead to inaccurate conclusions, and is incremental as it builds on existing SBI methods by testing them under new conditions.
The study investigated the impact of model misspecification on neural simulation-based inference (SBI) methods, finding that it can severely degrade performance, with no tested mitigation strategies preventing failure in all cases.
Aided by advances in neural density estimation, considerable progress has been made in recent years towards a suite of simulation-based inference (SBI) methods capable of performing flexible, black-box, approximate Bayesian inference for stochastic simulation models. While it has been demonstrated that neural SBI methods can provide accurate posterior approximations, the simulation studies establishing these results have considered only well-specified problems -- that is, where the model and the data generating process coincide exactly. However, the behaviour of such algorithms in the case of model misspecification has received little attention. In this work, we provide the first comprehensive study of the behaviour of neural SBI algorithms in the presence of various forms of model misspecification. We find that misspecification can have a profoundly deleterious effect on performance. Some mitigation strategies are explored, but no approach tested prevents failure in all cases. We conclude that new approaches are required to address model misspecification if neural SBI algorithms are to be relied upon to derive accurate scientific conclusions.