Uncertainty in Neural Processes
This work addresses the challenge of uncertainty estimation in probabilistic models for low-data scenarios, which is incremental as it builds on existing neural process frameworks.
The paper tackles the problem of improving posterior predictive inference in neural processes when conditioning data is scarce, demonstrating qualitative and quantitative improvements through specific architecture and objective choices, with superior results shown in image completion experiments.
We explore the effects of architecture and training objective choice on amortized posterior predictive inference in probabilistic conditional generative models. We aim this work to be a counterpoint to a recent trend in the literature that stresses achieving good samples when the amount of conditioning data is large. We instead focus our attention on the case where the amount of conditioning data is small. We highlight specific architecture and objective choices that we find lead to qualitative and quantitative improvement to posterior inference in this low data regime. Specifically we explore the effects of choices of pooling operator and variational family on posterior quality in neural processes. Superior posterior predictive samples drawn from our novel neural process architectures are demonstrated via image completion/in-painting experiments.