论文标题
解除武器：二进制潜在变量的对立梯度估计器
DisARM: An Antithetic Gradient Estimator for Binary Latent Variables
论文作者
论文摘要
由于难以准确估算梯度，具有离散潜在变量的训练模型具有挑战性。最近的许多进展是通过利用连续放松的系统来实现的，这些系统并不总是可用，甚至可能。增强 - 增强型 - 合并（ARM）估计器提供了一种替代方案，而不是放松，而是使用持续的增强。在增强变量上应用对立抽样会产生相对较低的变化和无偏估计器，适用于任何具有二进制潜在变量的模型。但是，尽管对立抽样降低了方差，但增强过程会增加方差。我们表明，可以通过分析整合通过增强过程引入的随机性来改善ARM，从而确保降低大量方差。我们的估计器解除武装非常易于实施，并且具有与ARM相同的计算成本。我们评估了几种生成建模基准的武装，并表明它始终优于ARM和强大的独立样品基线，从方差和对数可能性方面。此外，我们提出了一种本地版本的解除武器，旨在优化多样本变分界，并表明它的表现优于当前最新方法Vimco。
Training models with discrete latent variables is challenging due to the difficulty of estimating the gradients accurately. Much of the recent progress has been achieved by taking advantage of continuous relaxations of the system, which are not always available or even possible. The Augment-REINFORCE-Merge (ARM) estimator provides an alternative that, instead of relaxation, uses continuous augmentation. Applying antithetic sampling over the augmenting variables yields a relatively low-variance and unbiased estimator applicable to any model with binary latent variables. However, while antithetic sampling reduces variance, the augmentation process increases variance. We show that ARM can be improved by analytically integrating out the randomness introduced by the augmentation process, guaranteeing substantial variance reduction. Our estimator, DisARM, is simple to implement and has the same computational cost as ARM. We evaluate DisARM on several generative modeling benchmarks and show that it consistently outperforms ARM and a strong independent sample baseline in terms of both variance and log-likelihood. Furthermore, we propose a local version of DisARM designed for optimizing the multi-sample variational bound, and show that it outperforms VIMCO, the current state-of-the-art method.
