PriorGuide: Test-Time Prior Adaptation for Simulation-Based Inference
This addresses a problem for researchers in computational fields like engineering or neuroscience by allowing flexible updates to prior knowledge in pre-trained inference models, though it is incremental as it builds on existing diffusion-based methods.
The paper tackles the limitation of amortized simulator-based inference methods being constrained by the prior distribution used during training, and introduces PriorGuide, a technique that enables adaptation of trained diffusion models to new priors at test time without retraining, enhancing model versatility.
Amortized simulator-based inference offers a powerful framework for tackling Bayesian inference in computational fields such as engineering or neuroscience, increasingly leveraging modern generative methods like diffusion models to map observed data to model parameters or future predictions. These approaches yield posterior or posterior-predictive samples for new datasets without requiring further simulator calls after training on simulated parameter-data pairs. However, their applicability is often limited by the prior distribution(s) used to generate model parameters during this training phase. To overcome this constraint, we introduce PriorGuide, a technique specifically designed for diffusion-based amortized inference methods. PriorGuide leverages a novel guidance approximation that enables flexible adaptation of the trained diffusion model to new priors at test time, crucially without costly retraining. This allows users to readily incorporate updated information or expert knowledge post-training, enhancing the versatility of pre-trained inference models.