Apple's colorful mainstream flagship, starting at $749.
Apple's AI Team Working on LiDAR, Smaller Neural Networks, Image Processing and More
At an exclusive invitation-only lunch at an AI conference in Barcelona, Spain that took place on December 6, Apple head of machine learning Russ Salakhutdinov and other Apple employees talked about their work on artificial intelligence.
Quartz has obtained some of the slides that Salakhutdinov used during the presentation, giving us a glimpse at the advancements Apple is making in AI and machine learning.
Based on one of the slides, topics at the lunch were wide-ranging, including health and vital signs, volumetric detection of LiDAR, prediction with structured outputs, image processing and colorization, intelligent assistant and language modeling, and activity recognition.
One of the slides, related to Apple's LiDAR work, featured a picture of two cars, but Apple engineers did not mention cars or any automotive research, such as self-driving vehicles. Another slide focused on Apple's image recognition algorithms, which are able to process 3,000 images per second, twice as fast as Google's capabilities. Apple's work on smaller neural networks that can run directly on devices was also covered.
- Deep generative models
- Model compression
- Holistic scene understanding
- Model reliability
- Deep reinforcement learning
- Unsupervised learning, transfer learning, one-shot learning
- Reasoning, attention & memory
- Efficient training on distributed computing
Going forward, Apple plans to allow its AI and machine learning researchers to start publishing papers, marking its willingness to contribute to the research community. Apple has always been notoriously secretive, but allowing researchers to publish could attract top talent that would not otherwise want to join the company.
A full rundown of the slides from Apple's presentation can be seen over at Quartz.
Quartz has obtained some of the slides that Salakhutdinov used during the presentation, giving us a glimpse at the advancements Apple is making in AI and machine learning.
Based on one of the slides, topics at the lunch were wide-ranging, including health and vital signs, volumetric detection of LiDAR, prediction with structured outputs, image processing and colorization, intelligent assistant and language modeling, and activity recognition.
One of the slides, related to Apple's LiDAR work, featured a picture of two cars, but Apple engineers did not mention cars or any automotive research, such as self-driving vehicles. Another slide focused on Apple's image recognition algorithms, which are able to process 3,000 images per second, twice as fast as Google's capabilities. Apple's work on smaller neural networks that can run directly on devices was also covered.
Another slide focused on Apple's ability to build neural networks that are 4.5 times smaller than the originals with no loss in accuracy, and twice the speed. The technique, not unknown in AI research, uses a larger, more robust neural network to teach another network the decisions it would make in a variety of situations. The "student" network then has a streamlined version of the "teacher" network's knowledge. In essence, it predicts the larger network's predictions about a given photo or audio sample.The discussion also covered some of Apple's research focus areas:
- Deep generative models
- Model compression
- Holistic scene understanding
- Model reliability
- Deep reinforcement learning
- Unsupervised learning, transfer learning, one-shot learning
- Reasoning, attention & memory
- Efficient training on distributed computing
Going forward, Apple plans to allow its AI and machine learning researchers to start publishing papers, marking its willingness to contribute to the research community. Apple has always been notoriously secretive, but allowing researchers to publish could attract top talent that would not otherwise want to join the company.
A full rundown of the slides from Apple's presentation can be seen over at Quartz.
[ 43 comments ]
Top Rated Comments
(View all)
24 months ago
Going forward, Apple plans to allow its AI and machine learning researchers to start publishing papers, marking its willingness to contribute to the research community.
This is probably one of the more significant things I got from this post. Good on Apple and about time.
24 months ago
How about assigning one or two of your bazillion employees to a forward thinking new project -- a powerful i7-based desktop computer with modern parts and the ability to upgrade? It should take about two weeks...
24 months ago
First output from the AI neural network: Apple's laptops are too thin and need larger batteries. :)
24 months ago
But... But... But... Everyone told me Apple lost the AI race and they've given up
The AI race is just getting started.But like the OS war, browser war and Mobile war, they'll all end up looking almost the same.
24 months ago
I used a search tool, but i can't find the words "SkyNet" or "Terminator" in their presentation.
24 months ago
But... But... But... All of the Apple haters told me Apple lost the AI race and they've given up
24 months ago
Somebody's keen to get out some counter-narratives this season.
Meanwhile in my inbox this week I see Amazon have launched all manner of AI services on AWS - image recognition, speech recognition, natural language processing, AI conversation handling...all available to barely-competent coders like me.
Meanwhile in my inbox this week I see Amazon have launched all manner of AI services on AWS - image recognition, speech recognition, natural language processing, AI conversation handling...all available to barely-competent coders like me.
[ Read All Comments ]
Apple in macOS Mojave introduced Dynamic Desktops, which are wallpapers that shift with the time of day, changing the lighting and look of the wallpaper with the progress of the sun across the sky.
...
In macOS Mojave, Apple has introduced a range of new Finder Quick Actions that make it easier for you to perform quick edits to files without having to open the apps associated with them.
To...
In previous versions of macOS, the Quick Look feature lets you view photos and files without having to open them in an app. In macOS Mojave, Apple has also introduced some convenient new editing...
In macOS Mojave, Apple has extended its Continuity features so that now you can use your iPhone or iPad's camera to take a photo or scan a document, and it will be immediately available on your...
Apple's new macOS Mojave update is not compatible with mid-2010 and mid-2012 Mac Pros with stock GPUs, but it is supported on 2010 and 2012 Mac Pro models that have been upgraded with graphics...
In macOS Mojave, Apple has introduced a new screen capture interface that unifies the screenshot and screen recording features on Mac, making accessing them easier than it used to be.
A new...
If you're someone that has a lot of files on your desktop, you're going to love the new Stacks feature in macOS Mojave, which is designed to organize all of your files into neat little piles...
Apple today shared a support document that will be of interest to customers who own a late 2012 27-inch iMac with a 3TB hard drive and are attempting to upgrade to the macOS Mojave update.
There...
