Misspecification-robust Sequential Neural Likelihood for Simulation-based Inference
This work addresses a critical issue in simulation-based inference for researchers using mechanistic models, offering a robust solution to model misspecification, though it is incremental as it builds upon existing SNL methods.
The paper tackles the problem of unreliable performance and overconfident posteriors in sequential neural likelihood (SNL) methods under model misspecification in simulation-based inference, proposing a novel SNL method that incorporates adjustment parameters to achieve robustness, resulting in more accurate point estimates and uncertainty quantification compared to standard SNL.
Simulation-based inference techniques are indispensable for parameter estimation of mechanistic and simulable models with intractable likelihoods. While traditional statistical approaches like approximate Bayesian computation and Bayesian synthetic likelihood have been studied under well-specified and misspecified settings, they often suffer from inefficiencies due to wasted model simulations. Neural approaches, such as sequential neural likelihood (SNL) avoid this wastage by utilising all model simulations to train a neural surrogate for the likelihood function. However, the performance of SNL under model misspecification is unreliable and can result in overconfident posteriors centred around an inaccurate parameter estimate. In this paper, we propose a novel SNL method, which through the incorporation of additional adjustment parameters, is robust to model misspecification and capable of identifying features of the data that the model is not able to recover. We demonstrate the efficacy of our approach through several illustrative examples, where our method gives more accurate point estimates and uncertainty quantification than SNL.