Timezone: »

 
Poster
Compositional Score Modeling for Simulation-Based Inference
Tomas Geffner · George Papamakarios · Andriy Mnih

Tue Jul 25 05:00 PM -- 06:30 PM (PDT) @ Exhibit Hall 1 #801
in Poster Session 2 »

Neural Posterior Estimation methods for simulation-based inference can be ill-suited for dealing with posterior distributions obtained by conditioning on multiple observations, as they tend to require a large number of simulator calls to learn accurate approximations. In contrast, Neural Likelihood Estimation methods can handle multiple observations at inference time after learning from individual observations, but they rely on standard inference methods, such as MCMC or variational inference, which come with certain performance drawbacks. We introduce a new method based on conditional score modeling that enjoys the benefits of both approaches. We model the scores of the (diffused) posterior distributions induced by individual observations, and introduce a way of combining the learned scores to approximately sample from the target posterior distribution. Our approach is sample-efficient, can naturally aggregate multiple observations at inference time, and avoids the drawbacks of standard inference methods.

[ PDF] 

Author Information

Tomas Geffner (University of Massachusetts, Amherst)
George Papamakarios (DeepMind)
Andriy Mnih (DeepMind)

More from the Same Authors