Doubly Stochastic Variational Inference for Neural Processes with Hierarchical Latent Variables
Doubly Stochastic Variational Inference for Neural Processes with Hierarchical Latent Variables
复制标题
DOI：
--
复制
发表时间：
2020-07
期刊：
影响因子：
--
通讯作者：
Q. Wang;H. V. Hoof
中科院分区：
文献类型：
--
作者：
Q. Wang;H. V. Hoof
神经过程（Neural processes, NPs）是一类随机过程的变分近似模型，在计算效率和不确定性量化方面具有良好的性能。这些过程使用具有潜在变量输入的神经网络来诱导预测分布。然而，香草NPs的表达能力有限，因为它们只使用全局潜在变量，而目标特定的局部变异有时可能至关重要。为了应对这一挑战，我们系统地研究了NP模型，并提出了一种新的NP模型变体，我们称之为双重随机变分神经过程（dsnp）。该模型结合全局潜变量和局部潜变量进行预测。我们在几个实验中对该模型进行了评估，结果表明该模型在多输出回归和分类中的不确定性估计方面具有竞争力的预测性能。
Neural processes (NPs) constitute a family of variational approximate models for stochastic processes with promising properties in computational efficiency and uncertainty quantification. These processes use neural networks with latent variable inputs to induce predictive distributions. However, the expressiveness of vanilla NPs is limited as they only use a global latent variable, while target specific local variation may be crucial sometimes. To address this challenge, we investigate NPs systematically and present a new variant of NP model that we call Doubly Stochastic Variational Neural Process (DSVNP). This model combines the global latent variable and local latent variables for prediction. We evaluate this model in several experiments, and our results demonstrate competitive prediction performance in multi-output regression and uncertainty estimation in classification.