Nested Variational Inference
Nested Variational Inference
复制标题
DOI：
--
复制
发表时间：
2021-06
期刊：
影响因子：
--
通讯作者：
Heiko Zimmermann;Hao Wu;Babak Esmaeili;Sam Stites;Jan-Willem van de Meent
中科院分区：
文献类型：
--
作者：
Heiko Zimmermann;Hao Wu;Babak Esmaeili;Sam Stites;Jan-Willem van de Meent
我们开发了嵌套变分推理（NVI），这是一系列方法，通过最小化每个嵌套级别的前向或反向 KL 散度来学习嵌套重要性采样器的建议。 NVI适用于许多常用的重要性采样策略，并提供了一种学习中间密度的机制，可以作为启发式方法来指导采样器。我们的实验将 NVI 应用于（a）使用学习的退火路径从多模态分布中获取样本（b）学习启发式算法来近似隐马尔可夫模型中未来观察的可能性，以及（c）在分层深度生成模型中执行摊销推理。我们观察到，优化嵌套目标可以提高对数平均权重和有效样本量方面的样本质量。
We develop nested variational inference (NVI), a family of methods that learn proposals for nested importance samplers by minimizing an forward or reverse KL divergence at each level of nesting. NVI is applicable to many commonly-used importance sampling strategies and provides a mechanism for learning intermediate densities, which can serve as heuristics to guide the sampler. Our experiments apply NVI to (a) sample from a multimodal distribution using a learned annealing path (b) learn heuristics that approximate the likelihood of future observations in a hidden Markov model and (c) to perform amortized inference in hierarchical deep generative models. We observe that optimizing nested objectives leads to improved sample quality in terms of log average weight and effective sample size.