Letters
How to Manage Uncertainty
There’s a lot we don’t know about the future: When will a Covid-19 vaccine be available? Who will win the next election? Or in a business context, how many customers will we have next year?
Latest
The Batch Newsletter
The Batch: Deepfakes Against Oppression, Alexa Hears With Her Eyes, New Medical AI Standards, Action Recognition With a Twist
There’s a lot we don’t know about the future: When will a Covid-19 vaccine be available? Who will win the next election? Or in a business context, how many customers will we have next year?
Science
Prelude to a Quake?
Geologists call them slow slips: deep, low-frequency earthquakes that can last a month but have little effect on the surface. A model trained to predict such events could help with forecasting potentially catastrophic quakes.
Machine Learning Research
Style and Substance: An improved GAN technique for style transfer
GANs are adept at mapping the artistic style of one picture onto the subject of another, known as style transfer. However, applied to the fanciful illustrations in children’s books, some GANs prove better at preserving style, others better at preserving subject matter.
Machine Learning Research
The Telltale Artifact: A technique for detecting GAN-generated deepfakes
Deepfakes have gone mainstream, allowing celebrities to star in commercials without setting foot in a film studio. A new method helps determine whether such endorsements — and other images produced by generative adversarial networks — are authentic.
Science
More Data for Medical AI: AI recognizes medical scans without iodine dye.
Convolutional neural networks are good at recognizing disease symptoms in medical scans of patients who were injected with iodine-based dye, that makes their organs more visible. But some patients can’t take the dye. Now synthetic scans from a GAN are helping CNNs learn to analyze undyed images.
Machine Learning Research
Making GANs More Inclusive: A technique to help GANs represent their datasets fairly
A typical GAN’s output doesn’t necessarily reflect the data distribution of its training set. Instead, GANs are prone to modeling the majority of the training distribution, sometimes ignoring rare attributes — say, faces that represent minority populations.
Tech & Society
A Good Look for AI: How L’Oreal uses AI to power its virtual makeup try-on
Trying on new makeup is a hassle — apply, inspect, wash off, repeat. Not to mention the tribulation of visiting brick-and-mortar stores during the pandemic. Augmented reality is helping people try on all the makeup they want without leaving home.
Letters
Introducing the Generative Adversarial Networks Specialization
This special issue of The Batch celebrates the launch of our new Generative Adversarial Networks Specialization! GANs are among the most exciting technologies to emerge from deep learning.
The Batch Newsletter
The Batch: GAN Special Issue! Ian Goodfellow For Real, Detecting Fakes, Including Minorities, Synthesizing Training Data
This special issue of The Batch celebrates the launch of our new Generative Adversarial Networks Specialization! GANs are among the most exciting technologies to emerge from deep learning. These networks learn in a very different way than typical supervised methods...
Interviews & Essays
Ian Goodfellow: A Man, A Plan, A GAN
Brilliant ideas strike at unlikely moments. Ian Goodfellow conceived generative adversarial networks while spitballing programming techniques with friends at a bar. Goodfellow, who views himself as “someone who works on the core technology, not the applications,”...
Machine Learning Research
Guess What Happens Next: Research teaches robots to predict unseen obstacles.
New research teaches robots to anticipate what’s coming rather than focusing on what’s right in front of them. Researchers developed Occupancy Anticipation (OA), a navigation system that predicts unseen obstacles in addition to observing those in its field of view.