Symmetry-Aware Marginal Density Estimation
This work addresses scalability issues for researchers and practitioners in probabilistic modeling, particularly for statistical relational models, though it appears incremental as it builds on existing Rao-Blackwell theory.
The paper tackled the problem of scalable inference in large probabilistic models with symmetries by introducing a novel marginal density estimator, which was shown to outperform standard estimators by several orders of magnitude in empirical results.
The Rao-Blackwell theorem is utilized to analyze and improve the scalability of inference in large probabilistic models that exhibit symmetries. A novel marginal density estimator is introduced and shown both analytically and empirically to outperform standard estimators by several orders of magnitude. The developed theory and algorithms apply to a broad class of probabilistic models including statistical relational models considered not susceptible to lifted probabilistic inference.