arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2101.04141cs.LG
收藏

从修补到工程:Tensorflow Playground中的测量

From Tinkering to Engineering: Measurements in Tensorflow Playground

University of California, Berkeley(加州大学伯克利分校);NTNU(挪威科技大学)

Henrik Hoeiness, Axel Harstad, Gerald Friedland

摘要:

在本文中,我们将介绍Tensorflow Playground的一个扩展,称为Tensorflow Meter(简称TFMeter)。TFMeter是一个交互式神经网络架构工具,允许可视化创建神经网络的不同架构。除了它的祖先,操场,我们的工具显示信息理论的测量,同时构建,训练和测试网络。因此,每一个变化都会导致至少一个测量值的变化,从而为不同架构能够学习什么提供更好的工程直觉。测量值来自文献中的各个地方。在这个演示中,我们描述了我们的Web应用程序,该应用程序可在http://tfmeter.icsi.berkeley.edu/上在线获得,并认为与原始Playground旨在建立关于神经网络的直觉一样,我们的扩展可以教育用户可用的测量,我们希望最终能够改善该领域的实验设计和再现性。

英文摘要:

In this article, we present an extension of the Tensorflow Playground, called Tensorflow Meter (short TFMeter). TFMeter is an interactive neural network architecting tool that allows the visual creation of different architectures of neural networks. In addition to its ancestor, the playground, our tool shows information-theoretic measurements while constructing, training, and testing the network. As a result, each change results in a change in at least one of the measurements, providing for a better engineering intuition of what different architectures are able to learn. The measurements are derived from various places in the literature. In this demo, we describe our web application that is available online at http://tfmeter.icsi.berkeley.edu/ and argue that in the same way that the original Playground is meant to build an intuition about neural networks, our extension educates users on available measurements, which we hope will ultimately improve experimental design and reproducibility in the field.

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。

↑