WHY THIS MATTERS IN BRIEF
- AI’s that can analyse data in real time will become increasingly important as we come to rely on them more and more to help us make, and automate, decisions
It’s probably fair to say that in many cases, whether it’s the new breed of Personal Digital Assistants, or in the healthcare and security industries particularly, that you want your Artificial Intelligence (AI) to work in real time, on data as it happens. And Microsoft knows this as well.
This week the software juggernaut, who recently bought us AI’s that have learnt how to create their own programs, and wrangle pigs to make them better at cooperating, just unveiled its new hardware acceleration platform, Project Brainwave, that promises speedy, real time AI in the cloud.
Thanks to Intel’s new Stratix 10 Field Programmable Gate Array (FPGA) chip, which Microsoft have been using to turn their Azure cloud platform into the world’s biggest supercomputer, their new AI can crunch a hefty 39.5 teraflops in machine learning tasks with less than 1 millisecond of latency, without having to batch tasks together. In other words it can handle complex AI tasks as they’re received, and that’s huge.
The new platform is considerably more flexible than many of its hard coded rivals, too, and it relies on a “soft” Dynamic Neural Network (DNN) processing engine that’s just “dropped” into the new FPGA’s whereas most of Microsoft’s competitors often have to lock their AI’s into their FPGA’s from the outset and that limits their flexibility and performance.
The new platform is also integrated with Microsoft’s own AI framework, called Cognitive Toolkit, and works with a variety of other AI platforms including Google’s TensorFlow, an open approach gives users the flexibility to build new machine learning systems the way they want to while giving them the advantage of letting them run them all in real time, instead of letting the hardware dictate their methods.
To no one’s surprise, Microsoft plans to make Project Brainwave available via its own Azure cloud services so companies can make use of live AI, and while there’s no guarantee that it’s going to be widely adopted it’s a bold and important step forwards, and one that Microsoft hopes will give it an edge over its rivals in what’s becoming an increasingly competitive market.
