A Universal Marginalizer for Amortized Inference in Generative Models
This work addresses inference challenges in generative models for researchers and practitioners, offering an incremental improvement in efficiency.
The authors tackled the problem of varying observation sets in causal generative models by training a single neural network to approximate all conditional marginal distributions, amortizing inference costs and improving importance sampling efficiency.
We consider the problem of inference in a causal generative model where the set of available observations differs between data instances. We show how combining samples drawn from the graphical model with an appropriate masking function makes it possible to train a single neural network to approximate all the corresponding conditional marginal distributions and thus amortize the cost of inference. We further demonstrate that the efficiency of importance sampling may be improved by basing proposals on the output of the neural network. We also outline how the same network can be used to generate samples from an approximate joint posterior via a chain decomposition of the graph.