Conditional Neural Processes

TL;DR

Conditional Neural Processes (CNPs) are neural models that combine the flexibility of Gaussian Processes with the scalability of neural networks, enabling quick and accurate predictions with limited data across various tasks.

Contribution

The paper introduces Conditional Neural Processes, a new neural network family that integrates the advantages of GPs and neural networks, trained via gradient descent for efficient function approximation.

Findings

CNPs achieve accurate predictions with few data points.

CNPs scale to complex functions and large datasets.

CNPs perform well on regression, classification, and image completion tasks.

Abstract

Deep neural networks excel at function approximation, yet they are typically trained from scratch for each new function. On the other hand, Bayesian methods, such as Gaussian Processes (GPs), exploit prior knowledge to quickly infer the shape of a new function at test time. Yet GPs are computationally expensive, and it can be hard to design appropriate priors. In this paper we propose a family of neural models, Conditional Neural Processes (CNPs), that combine the benefits of both. CNPs are inspired by the flexibility of stochastic processes such as GPs, but are structured as neural networks and trained via gradient descent. CNPs make accurate predictions after observing only a handful of training data points, yet scale to complex functions and large datasets. We demonstrate the performance and versatility of the approach on a range of canonical machine learning tasks, including regression, classification and image completion.