News

March 23, 2019

Humans Can Predict How Machines (Mis)classify Adversarial Images

How could AI label an image of a dog as a pineapple? Research shows there’s a logic to these errors that humans can intuitively understand.

March 22, 2019

Giant Steps and Liberating Spaces – Virtual Reality is Making Cool Moves

VR researchers are liberating users from the small, object-free player settings of today, to allow them to safely roam the real world.

March 22, 2019

MIT Develops Algorithm to Accelerate Neural Networks by 200x

AI researchers have found a method of identifying likely neural network architectures that are much faster than previous methods.

March 21, 2019

NYU Open-sources Breast Cancer Screening Model Trained on Over 200,000 Mammography Exams

Researchers at NYU have open-sourced a machine learning model they say can predict tumor malignancy in breasts with high accuracy.

March 21, 2019

TensorFlow 2.0 and Cloud AI Make It Easy to Train, Deploy, and Maintain Scalable Machine Learning Models

TensorFlow has matured into an entire end-to-end ML ecosystem that includes a variety of tools, libraries, and deployment options.

March 20, 2019

Reducing the Need for Labeled Data in Generative Adversarial Networks

Generative adversarial networks (GANs) are a powerful class of deep generative models.

March 19, 2019

Google’s Stadia Uses Style Transfer ML to Manipulate Video Game Environments

Style transfer is a machine learning technique that recomposes images in the aesthetic of other images.

March 19, 2019

Measuring the Limits of Data Parallel Training for Neural Networks

Neural networks have achieved state-of-the-art results in a wide variety of prediction tasks, including image classification.

March 19, 2019

Stroke of Genius: GauGAN Turns Doodles into Stunning, Photorealistic Landscapes

March 18, 2019 - A novice painter might set brush to canvas aiming to create a stunning sunset landscape —  craggy, snow-covered peaks reflected in a glassy lake — only to end up with something that looks more like a multi-colored inkblot.
March 19, 2019

NVIDIA’s RAPIDS Joins Google's Set of Deep Learning VM Images for Faster Data Science

The Deep Learning VM Images comprise a set of Debian 9-based Compute Engine disk images optimized for data science & machine learning tasks.