Apple has launched a new online research site to share the machine learning posts written by Apple engineers. The
Apple Machine Learning Journal is first of its kind and as of now, it has only one post apart from the following Welcome message.
“Welcome to the Apple Machine Learning Journal. Here, you can read posts written by Apple engineers about their work using machine learning technologies to help build innovative products for millions of people around the world. If you’re a machine learning researcher or student, an engineer or developer, we’d love to hear your questions and feedback. “
The launch of this new online journal is a bit surprising for all because Apple has always remained secretive about its research projects, even after contributing to some important open-source projects, like Webkit and Swift.
The
first post on this research blog is based on “Improving the Realism of Synthetic Images” that literally means training the neural network by turning synthetic images into realistic ones. This post is based on a research paper that was first published on arXiv, but the language in the here is much more simplified, maybe because Apple wants it to be read by a large number of developers worldwide.
Source: machinelearning.apple.com
In this post, Apple has described how they train the neural network to detect the objects, like faces, in a given picture. Here, Apple has used a synthetic image with computer-generated characters instead of using millions of sample photos. Well, this is smart because it is a cheaper and faster way to train the neural network. The GIF illustrations are also added to support the shared content, which make the post better understandable.
Source: Apple
Seeing this move of the “very secretive” company, we can say that Apple has now started opening up with the world. This new move shows that the company is working on bridging the gap that it has created in these years. And, this not only helps researchers and engineers working in this domain but it will help the company as well because now, with direct interaction with scholars, they will get honest feedback and thousands of different kinds of queries, which eventually will need more brainstorming and lead towards a better outcome.
What we see here is a better opportunity for developers, engineers and researchers, who are working on Machine Learning. Now, they can give the right direction to their projects with the help of a company that has shown proven results in past.