Bayesian Models of Data Streams with Hierarchical Power Priors
This work addresses the challenge of data stream analysis for applications requiring real-time inference, though it appears incremental as it builds on existing Bayesian and variational methods.
The paper tackled the problem of making inferences from data streams by addressing continuous model updating and adapting to changes in data distribution, proposing a Bayesian approach with hierarchical priors and a novel variational inference scheme, validated on three real datasets.
Making inferences from data streams is a pervasive problem in many modern data analysis applications. But it requires to address the problem of continuous model updating and adapt to changes or drifts in the underlying data generating distribution. In this paper, we approach these problems from a Bayesian perspective covering general conjugate exponential models. Our proposal makes use of non-conjugate hierarchical priors to explicitly model temporal changes of the model parameters. We also derive a novel variational inference scheme which overcomes the use of non-conjugate priors while maintaining the computational efficiency of variational methods over conjugate models. The approach is validated on three real data sets over three latent variable models.