Beyond ELBOs: A Large-Scale Evaluation of Variational Methods for Sampling
This work addresses a practical problem for researchers and practitioners in machine learning by offering a comprehensive benchmark to assess and compare sampling methods, though it is incremental as it builds on existing evaluation efforts.
The authors tackled the lack of a unified evaluation framework for sampling methods like Monte Carlo and Variational Inference by introducing a benchmark with standardized tasks and performance criteria, including novel metrics for mode collapse, to provide insights into method strengths and weaknesses.
Monte Carlo methods, Variational Inference, and their combinations play a pivotal role in sampling from intractable probability distributions. However, current studies lack a unified evaluation framework, relying on disparate performance measures and limited method comparisons across diverse tasks, complicating the assessment of progress and hindering the decision-making of practitioners. In response to these challenges, our work introduces a benchmark that evaluates sampling methods using a standardized task suite and a broad range of performance criteria. Moreover, we study existing metrics for quantifying mode collapse and introduce novel metrics for this purpose. Our findings provide insights into strengths and weaknesses of existing sampling methods, serving as a valuable reference for future developments. The code is publicly available here.