Principled Parallel Mean-Field Inference for Discrete Random Fields
This work addresses a bottleneck in inference for probabilistic models, offering a more efficient and robust method for researchers and practitioners in machine learning, though it is incremental as it builds on existing variational inference frameworks.
The paper tackles the inefficiency and impracticality of standard mean-field variational inference in discrete random fields by proposing a novel proximal gradient-based optimization approach, which demonstrates faster convergence and often finds better optima than traditional techniques, with reduced sensitivity to parameter choices.
Mean-field variational inference is one of the most popular approaches to inference in discrete random fields. Standard mean-field optimization is based on coordinate descent and in many situations can be impractical. Thus, in practice, various parallel techniques are used, which either rely on ad-hoc smoothing with heuristically set parameters, or put strong constraints on the type of models. In this paper, we propose a novel proximal gradient-based approach to optimizing the variational objective. It is naturally parallelizable and easy to implement. We prove its convergence, and then demonstrate that, in practice, it yields faster convergence and often finds better optima than more traditional mean-field optimization techniques. Moreover, our method is less sensitive to the choice of parameters.