arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1612.04799stat.MLcs.CVcs.LG
收藏

Deep Function Machines:用于拓扑层表达的广义神经网络

Deep Function Machines: Generalized Neural Networks for Topological Layer Expression

Machine Learning at Berkeley(伯克利机器学习);University of California, Berkeley(加州大学伯克利分校)

William H. Guss

摘要:

在本文中,我们提出了一种广义的深度神经网络,称为深度函数机(DFM)。DFM作用于任意(可能是无限)维的向量空间,我们证明了一族DFM对输入数据的维数是不变的;也就是说,模型的参数化并不直接取决于输入的质量(例如:高分辨率图像)。利用这一推广,我们提供了一个新的理论的泛逼近的有界非线性算子之间的函数空间。然后,我们建议DFMs提供了一个表达框架,设计新的神经网络层类型的拓扑考虑。最后,我们介绍了一种新的架构,RippLeNet,分辨率不变的计算机视觉,凭经验达到最先进的不变性。

英文摘要:

In this paper we propose a generalization of deep neural networks called deep function machines (DFMs). DFMs act on vector spaces of arbitrary (possibly infinite) dimension and we show that a family of DFMs are invariant to the dimension of input data; that is, the parameterization of the model does not directly hinge on the quality of the input (eg. high resolution images). Using this generalization we provide a new theory of universal approximation of bounded non-linear operators between function spaces. We then suggest that DFMs provide an expressive framework for designing new neural network layer types with topological considerations in mind. Finally, we introduce a novel architecture, RippLeNet, for resolution invariant computer vision, which empirically achieves state of the art invariance.

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。

↑