arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1807.01613cs.LGstat.ML
收藏

条件神经过程

Conditional Neural Processes

DeepMind(深度思维);Imperial College London(帝国理工学院)

Marta Garnelo, Dan Rosenbaum, Chris J. Maddison, Tiago Ramalho, David Saxton, Murray Shanahan, Yee Whye Teh, Danilo J. Rezende, S. M. Ali Eslami

摘要:

深度神经网络擅长于函数逼近,但它们通常是为每个新函数从头开始训练的。另一方面,贝叶斯方法,如高斯过程(GP),利用先验知识在测试时快速推断新函数的形状。然而,GP在计算上是昂贵的,并且很难设计适当的先验。在本文中,我们提出了一系列的神经模型,条件神经过程(CNP),结合两者的好处。CNP的灵感来自于随机过程(如GP)的灵活性,但它被构造为神经网络,并通过梯度下降进行训练。CNP在仅观察少数训练数据点后就能做出准确的预测,但可以扩展到复杂的函数和大型数据集。我们展示了一系列典型的机器学习任务,包括回归,分类和图像完成的方法的性能和多功能性。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑