Direct-Manipulation Visualization of Deep Networks

TL;DR

TensorFlow Playground provides an interactive visualization tool that helps non-experts intuitively understand deep neural networks by allowing direct manipulation of hyperparameters and structures without coding.

Contribution

It introduces an open-source, interactive visualization platform that facilitates intuitive learning of deep networks for novices through direct manipulation.

Findings

Enables quick understanding of neural network behavior

Facilitates learning of hyperparameter effects

Supports non-expert exploration of deep learning concepts

Abstract

The recent successes of deep learning have led to a wave of interest from non-experts. Gaining an understanding of this technology, however, is difficult. While the theory is important, it is also helpful for novices to develop an intuitive feel for the effect of different hyperparameters and structural variations. We describe TensorFlow Playground, an interactive, open sourced visualization that allows users to experiment via direct manipulation rather than coding, enabling them to quickly build an intuition about neural nets.