Delivering on promise, Apple publishes its first artificial intelligence research paper

robot-507811_1280

Apple Inc. has made good on its promise to start publishing its research on artificial intelligence with the publication of its first paper in the field.

Spotted first by Forbes, the paper, titled “Learning from Simulated and Unsupervised Images through Adversarial Training,” was submitted for review in mid-November before being ultimately published by the Cornell University Library on Dec. 22.

The research covers the techniques required for training computer vision algorithms to recognize computer generated versus real world images. According to the research, synthetic image data is often “not realistic enough,” leading the network to learn details only present in synthetic images and fail to generalize well on real images.

The paper proposes that a solution to that problem involves “Simulated+Unsupervised learning,” a process that relies on a new machine learning technique. Generative Adversarial Networks are said to increase the realism of a simulated image by pitting two neural network, generator and discriminator, against each other to discern generated data from real data.

In delivering the outcome, Apple researchers have made key modifications to the standard GAN algorithm to preserve annotations, avoid artifacts and stabilize training, including a “self-regularization” term, local adversarial loss and updating the discriminator using a history of refined images. That led to enabling the generation of highly realistic images.

In more simple terms, the processes described in the research paper help computers see artificially generated images better through improved implementation of artificial intelligence and machine learning.

Apple is expected to publish more research in the year ahead in the fields of “volumetric detection of LiDAR” and “prediction of structured outputs,” both areas the company revealed it was researching earlier this month.

The research paper was written by Apple vision expert Ashish Shrivastava with the support of engineers including Tomas Pfister, Oncel Tuzel, Wenda Wang, Russ Webb and Apple Director of Artificial Intelligence Research Josh Susskind.

Image credit: Maxpixel/Public Domain CC0