GAN Lab: Understanding Complex Deep Generative Models using Interactive Visual Experimentation

• Published in: IEEE transactions on visualization and computer graphics Vol. 25; no. 1; pp. 310 - 320
• Main Authors: Kahng, Minsuk, Thorat, Nikhil, Chau, Duen Horng, Viegas, Fernanda B., Wattenberg, Martin
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.01.2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Data visualization; Deep learning; Experimentation; explorable explanations; Gallium nitride; Generative adversarial networks; Generators; information visualization; interactive experimentation; Machine learning; New technology; Training; visual analytics
• ISSN: 1077-2626; 1941-0506; 1941-0506
• DOI: 10.1109/TVCG.2018.2864500

Recent success in deep learning has generated immense interest among practitioners and students, inspiring many to learn about this new technology. While visual and interactive approaches have been successfully developed to help people more easily learn deep learning, most existing tools focus on simpler models. In this work, we present GAN Lab, the first interactive visualization tool designed for non-experts to learn and experiment with Generative Adversarial Networks (GANs), a popular class of complex deep learning models. With GAN Lab, users can interactively train generative models and visualize the dynamic training process's intermediate results. GAN Lab tightly integrates an model overview graph that summarizes GAN's structure, and a layered distributions view that helps users interpret the interplay between submodels. GAN Lab introduces new interactive experimentation features for learning complex deep learning models, such as step-by-step training at multiple levels of abstraction for understanding intricate training dynamics. Implemented using TensorFlow.js, GAN Lab is accessible to anyone via modern web browsers, without the need for installation or specialized hardware, overcoming a major practical challenge in deploying interactive tools for deep learning.