In this presentation and interview, Stéphane Cordova, VP of embedded technology at Kalray, provides an insight into the company’s artificial intelligence computing systems for autonomous cars, as well as Kalray itself. (See video below.)
Kalray recently unveiled its new neural network platform, which it calls “KaNN”, saying it enables “high-performance machine learning” applications for a variety of sectors including automotive, one of the areas in which it specialises.
KaNN and Kalray’s new AI processor, called the “Massively Parallel Processor Array”, or MPPA, also have potential applications in areas such as intelligent storage servers, data centers, robotics, drones, and others.
Speaking at the launch of the new technologies, Cordova says: “The Kalray MPPA intelligent processor is capable of processing massive amounts of data and trillions of instructions in parallel and in real time, making it the perfect architecture for Artificial Intelligence applications like self-driving cars and intelligent storage algorithms.
“Today, we are pleased to release the third version of the Kalray Neural Network platform to the AI and machine learning community. The KaNN is modular, compatible, easy-to-use and customizable, and it empowers developers to unleash the full potential and performance of AI applications built for Kalray MPPA processors.”
Coupled with KaNN, the MPPA processor enables developers to port frameworks such a variety of widely-used machine learning frameworks, such as Caffe, Torch and TensorFlow.
Kalray itself is a French company with offices in the US, and is what’s called a “fabless” semiconductor company, which means it contracts out the fabrication part of the process, and concentrates solely on designing the processors.
It’s backed by more than $60 million of investment, and holds patents in 23 different areas within microprocessor-related technologies.