Statistical Model Aggregation via Parameter Matching
This addresses the challenge of model aggregation for researchers and practitioners dealing with sequestered data, but it appears incremental as it builds on existing Bayesian nonparametric tools.
The paper tackles the problem of aggregating models from heterogeneous datasets by developing a meta-modeling framework that identifies correspondences among local parameters, and demonstrates its utility on tasks like text summarization and temperature forecasting with applications across various model types.
We consider the problem of aggregating models learned from sequestered, possibly heterogeneous datasets. Exploiting tools from Bayesian nonparametrics, we develop a general meta-modeling framework that learns shared global latent structures by identifying correspondences among local model parameterizations. Our proposed framework is model-independent and is applicable to a wide range of model types. After verifying our approach on simulated data, we demonstrate its utility in aggregating Gaussian topic models, hierarchical Dirichlet process based hidden Markov models, and sparse Gaussian processes with applications spanning text summarization, motion capture analysis, and temperature forecasting.