LGMLOct 26, 2022

Maximum Likelihood Learning of Unnormalized Models for Simulation-Based Inference

arXiv:2210.14756v24 citationsh-index: 89
Originality Incremental advance
AI Analysis

This work addresses simulation-based inference for experimental observations, offering a more efficient alternative to existing methods, though it appears incremental as it builds on synthetic likelihood approaches with a novel combination of EBMs and KL loss.

The authors tackled the problem of simulation-based inference (SBI) by introducing two synthetic likelihood methods that learn conditional energy-based models (EBMs) using synthetic data from a simulator, enabling posterior estimation with MCMC. They demonstrated improved performance on a neuroscience model, outperforming prior methods with a fraction of the simulation budget.

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.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes