Efficient Localized Inference for Large Graphical Models
This work addresses computational bottlenecks for researchers and practitioners dealing with large-scale probabilistic inference, though it appears incremental as it builds on existing correlation decay properties and error bounds.
The paper tackles the problem of efficiently approximating marginal distributions in large graphical models by proposing a localized inference algorithm that constructs a smaller model around the query variable, achieving fast and accurate results as verified on various datasets.
We propose a new localized inference algorithm for answering marginalization queries in large graphical models with the correlation decay property. Given a query variable and a large graphical model, we define a much smaller model in a local region around the query variable in the target model so that the marginal distribution of the query variable can be accurately approximated. We introduce two approximation error bounds based on the Dobrushin's comparison theorem and apply our bounds to derive a greedy expansion algorithm that efficiently guides the selection of neighbor nodes for localized inference. We verify our theoretical bounds on various datasets and demonstrate that our localized inference algorithm can provide fast and accurate approximation for large graphical models.