Wasserstein GAN

What Does Wasserstein GAN Mean?

The Wasserstein GAN (WGAN) is an algorithm introduced in a paper written by Martin Arjovsky, Soumith Chintala and Léon Bottou at the Courant Institute of Mathematical Sciences. The paper examines methods for unsupervised learning, and provides part of the roadmap for dealing with the pursuit of certain outcomes in machine learning projects.


Techopedia Explains Wasserstein GAN

The Wasserstein GAN algorithm is a variation of generative adversarial networks (GANs). Generative adversarial networks feature capabilities related to discriminating between data sets and choosing outcomes are fundamentally useful in machine learning. The Wasserstein GAN is a specific kind of GAN that, according to the team, “minimizes a reasonable and efficient approximation of the Earth Mover’s distance,” where the EM distance is a method to look at dissimilarity between two multidimensional data sets.

Through helping to deal with major training problems of generative adversarial networks in general, the Wasserstein GAN can be useful in the pursuit of dimensionality reduction and other goals related to specific machine learning outcomes.


Related Terms

Margaret Rouse
Technology Expert

Margaret is an award-winning technical writer and teacher known for her ability to explain complex technical subjects to a non-technical business audience. Over the past twenty years, her IT definitions have been published by Que in an encyclopedia of technology terms and cited in articles by the New York Times, Time Magazine, USA Today, ZDNet, PC Magazine, and Discovery Magazine. She joined Techopedia in 2011. Margaret's idea of a fun day is helping IT and business professionals learn to speak each other’s highly specialized languages.