拆解五因素影响，给出高效流变分推断的实用配方
Disentangling impact of capacity, objective, batchsize, estimators, and step-size on flow VI
- 逐项分析容量、目标函数等五大因素对性能的影响
- 提出可媲美甚至超越HMC的流变分推断方案
- 针对常见后验难题设计可精确采样的合成基准
基于归一化流的变分推断（flow VI）是一种有前景的近似推断方法，但其性能在不同研究中表现不一。众多算法选择会影响flow VI的表现。本文通过逐步分析，分离出容量、目标函数、梯度估计器、梯度估计次数（批大小）和步长这五个关键因素的影响。每一步均在控制其他变量的前提下，结合前序洞察与大规模并行计算进行评估。为实现高保真评估，我们构建了一个包含常见后验病态的合成基准，并支持精确采样。研究给出了各因素的具体建议，并提出一个可匹配或超越主流无迹哈密顿蒙特卡洛（HMC）方法的flow VI使用配方。
原文摘要 · Abstract (English)
Normalizing flow-based variational inference (flow VI) is a promising approximate inference approach, but its performance remains inconsistent across studies. Numerous algorithmic choices influence flow VI's performance. We conduct a step-by-step analysis to disentangle the impact of some of the key factors: capacity, objectives, gradient estimators, number of gradient estimates (batchsize), and step-sizes. Each step examines one factor while neutralizing others using insights from the previous steps and/or using extensive parallel computation. To facilitate high-fidelity evaluation, we curate a benchmark of synthetic targets that represent common posterior pathologies and allow for exact sampling. We provide specific recommendations for different factors and propose a flow VI recipe that matches or surpasses leading turnkey Hamiltonian Monte Carlo (HMC) methods.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。