Maximum Likelihood Learning of Unnormalized Models for Simulation-Based Inference

2022/10/26 by Glaser, Pierre, Arbel, Michael, Hromadka, Samo +2 · 3 citations
#FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · doi:10.48550/arxiv.2210.14756

Abstract

We introduce two synthetic likelihood methods for Simulation-Based Inference (SBI), to conduct either amortized or targeted inference from experimental observations when a high-fidelity simulator is available. Both methods learn a conditional energy-based model (EBM) of the likelihood using synthetic data generated by the simulator, conditioned on parameters drawn from a proposal distribution. The learned likelihood can then be combined with any prior to obtain a posterior estimate, from which samples can be drawn using MCMC. Our methods uniquely combine a flexible Energy-Based Model and the minimization of a KL loss: this is in contrast to other synthetic likelihood methods, which either rely on normalizing flows, or minimize score-based objectives; choices that come with known pitfalls. We demonstrate the properties of both methods on a range of synthetic datasets, and apply them to a neuroscience model of the pyloric network in the crab, where our method outperforms prior art for a fraction of the simulation budget.

Cited by

• Fast and Scalable Score-Based Kernel Calibration Tests
• Learning Latent Energy-Based Models via Interacting Particle Langevin Dynamics
• Uniform-in-time convergence bounds for Persistent Contrastive Divergence Algorithms

Related

• Simulation-Based Inference: A Practical Guide
• Variational methods for simulation-based inference
• Compositional simulation-based inference for time series
• Bayesian Optimization for Likelihood-Free Inference of Simulator-Based Statistical Models
• Simultaneous identification of models and parameters of scientific simulators