Word of the Day: generative adversarial network (GAN)

Word of the Day: Daily updates on the latest technology terms

Word of the Day Daily updates on the latest technology terms | August 21, 2019 generative adversarial network (GAN) A generative adversarial network (GAN) is a machine learning (ML) model in which two neural networks compete with each other to become more accurate in their predictions. GANs typically run unsupervised and use a cooperative zero-sum game framework to learn. The two neural networks that make up a GAN are referred to as the generator and the discriminator. The generator is a convolutional neural network and the discriminator is a deconvolutional neural network. The goal of the generator is to artificially manufacture outputs that could easily be mistaken for real data. The goal of the discriminator is to identify which outputs it receives have been artificially created. Essentially, GANs create theor own training data. As the feedback loop between the adversarial networks continue, the generator will begin to produce higher-quality output and the discriminator will become better at flagging data that has been artificially created. How GANs work The first step in establishing a GAN is to identify the desired end output and gather an initial training dataset based on those parameters. This data is then randomized and input into the generator until it acquires basic accuracy in producing outputs. After this, the generated images are fed into the discriminator along with actual data points from the original concept. The discriminator filters through the information and returns a probability between 0 and 1 to represent each image's authenticity (1 correlates with real and 0 correlates with fake). These values are then manually checked for success and repeated until the desired outcome is reached. Popular use cases for GANs   GANs are becoming a popular ML model for online retail sales because of their ability to understand and recreate visual content with increasingly remarkable accuracy.  Use cases include: Filling in images from an outline. Generating a realistic image from text. Producing photorealistic depictions of product prototypes. Converting black and white imagery into color.   In video production, GANs can be used to: Model patterns of human behavior and movement within a frame. Predict subsequent video frames. Create deepfake videos. Quote of the Day "Since GANs are capable of analyzing and recognizing detailed data, these systems are a powerhouse for generating artificial content." - Ronald Schmelzer Trending Terms adversarial machine learning convolutional neural network deconvolutional networks backpropagation algorithm deepfake zero-sum game co-opetition Learning Center 3 GAN use cases that showcase their positive potential GANs have notoriously been used to start a conversation around privacy, transparency and threats that AI can bring. Image generation GANs tap into the deepest fears of internet users: trickery. But GANs can also be a force for good, and blossoming GAN use cases show its true potential. Generative adversarial networks could be most powerful algorithm in AI Among all the different algorithms in AI, generative adversarial networks are 'the most interesting' development and could provide a path to more natural AI applications. New uses for GAN technology focus on optimizing existing tech If you're an internet user, you've likely seen GANs at work in the form of clipped-together videos of celebrities singing popular songs or delivering fake speeches. In more practical applications, GAN technology can safeguard robot learning and improve manufacturing processes. Enterprises need to plan for deepfake technology Deepfake technology can damage brands and compromise security. Combatting the threat won't be easy. This tutorial spells out the dangers and some steps enterprises can take now to protect themselves against deepfakes. How machine learning-powered password guessing impacts security Password guessing could be an even bigger threat to enterprises now that hackers can use machine learning to automate it. Here's what this means for security. Quiz Yourself Some employees fear that computers will take over their jobs with the _______ of machine learning, but that is not the case. A. raise B. rise Answer Stay in Touch For feedback about any of our definitions or to suggest a new definition, please contact me at: mrouse@techtarget.com Visit the Word of the Day Archives and catch up on what you've missed! FOLLOW US About This E-Newsletter The Word of the Day is published by TechTarget, Inc., 275 Grove Street, Newton, Massachusetts, 02466 US. Click to: Unsubscribe. You are receiving this email because you are a member of TechTarget. When you access content from this email, your information may be shared with the sponsors or future sponsors of that content and with our Partners, see up-to-date Partners List, as described in our Privacy Policy. For additional information, please contact: webmaster@techtarget.com. © 2019 TechTarget, Inc. all rights reserved. Designated trademarks, brands, logos, and service marks are the property of their respective owners. Privacy Policy | Partners List