Nvidia deep dives into deep learning with Pascal, Titan X GPUs

NVIDIA CEO and co-founder Jen-Hsun Huang unveiled several new technologies in line with its deep learning strategy amid the GPU Technology Conference in Silicon Valley on Tuesday.

zdnet-nvidiadigitsdeeplearninggputrainingsystem.jpg

Nvidia is hoping to pack the most punch it can into a single GPU with this month's debut of Titan X, which is being touted for cutting-edge mobile gaming and deep learning alike.

Artificial intelligence in your shopping basket: Machine learning for online retailers

AI techniques are becoming part of every day computing: here's how they're being used to help online retailers keep up with the competition.

Read More

But Titan X is just one of many moving parts Nvidia has in the works to get ahead in artificial, machine intelligence.

NVIDIA CEO and co-founder Jen-Hsun Huang unveiled several new technologies in line with its deep learning strategy amid the GPU Technology Conference in Silicon Valley on Tuesday.

Huang put the spotlight on the Pascal GPU series, which promise to speed up deep learning applications tenfold compared to Nvidia's current-generation Maxwell processors.

Even more pressure resides on Pascal thanks to its inclusion of 3D memory, which is also supposed to enable up to five times better performance in deep learning apps.

Pascal's objective is "mixed precision" computing, said to offer greater accuracy in getting computers to teach themselves.

First unveiled at the Game Developers Conference earlier this month, Titan X is at the center of the portfolio, boasted by its makers to be "most powerful processor ever built for training deep neural networks."

Based on the Maxwell GPU architecture, some of the suggested use cases for implementation on AI projects range from medical research to self-driving cars.

zdnet-geforcegtxtitanxtop.jpg
Tim Tabke / Phoenix Productions
But Titan might reach a broader audience -- and by extension -- spread more awareness about deep learning in general.

To spur development and deployment, Nvidia has developed specific graphics software to train machines on how to automatically recognize and classify objects with deep neural networks.

Dubbed Digits, the Deep Learning GPU Training System includes real-time monitoring and visualization features for building deep neural nets.

Nvidia also peeled back the curtain on an internal project its engineers have been using for deep learning research. The Digits DevBox is comprised of both Digits software and a quartet of Titan X GPUs -- not to mention several popular deep learning frameworks -- altogether of which promises up to four times faster development.

Titan X is now available, starting at $999. Pascal is scheduled to roll out next year.

Images via Nvidia

Newsletters

You have been successfully signed up. To sign up for more newsletters or to manage your account, visit the Newsletter Subscription Center.
See All
See All