Intel Dangles Machine Learning Chips on the Edge

Feb. 28, 2018
Intel Dangles Machine Learning Chips at the Edge

Last year, Intel embedded its line of Myriad 2 chips into a USB stick that developers could plug into development boards or personal computers to experiment with artificial intelligence. The custom chips can accelerate neural networks used in image recognition and other tricky tasks that computers can be trained to tackle.

On Tuesday, the company introduced a program to make it easier for developers to turn prototypes built with the Neural Compute Stick into security cameras, industrial sensors, and other production devices. Intel is partnering with Aaeon to offer a board called the A.I. Core, which serves as a sort of production version of the dongle.

The A.I. Core board contains the same Myriad vision processing unit as the stick, allowing small companies and entrepreneurs to move to the production board without changing code. Aaeon, a subsidiary of the computer hardware maker Asus, said that it would also offer custom board manufacturing as part of the program with Intel.

The program, called Intel AI: In Production, signals a step toward the wider deployment of custom chips for machine learning. Multinational chip makers like Intel and startups are targeting embedded chips that run inferencing algorithms without being connected to the cloud, reducing latency and bolstering security.

Most chip companies focused on machine learning are targeting data centers, where behemoths like Google and Amazon have amassed the computing power required for training and inferencing. Other companies, in a bid to shake Nvidia’s hold on the market for machine learning chips, are trying to conquer the edge.

Intel acquired its Myriad line in 2016 from the startup Movidius, which built the vision processing unit (VPU) to take machine learning tasks out of a main processor’s hands. With the acquisition of Movidius and server chip startup Nervana System, Intel is trying to stamp a stronger claim on the market for machine learning chips.

 “A.I. Core bridges the gap between the lab and volume production, allowing the innovators who adopted the Intel Movidius Neural Compute Stick to roll out a field deployment,” said Fabrizio Del Maffeo, Aaeon’s European vice president of technology, in a statement.

Intel said that the stick has been used by tens of thousands of developers since it was released seven months ago and its Myriad chips are used in several drones and other products. Google’s Clips home camera is embedded with a Myriad chip to make decisions about when to take photographs of a person’s family and friends.

About the Author

James Morra | Senior Staff Editor

James Morra is a senior staff editor for Electronic Design, where he covers the semiconductor industry and new technology trends. He also reports on the business behind electrical engineering, including the electronics supply chain. He joined Electronic Design in 2015 and is based in Austin, Texas.

Sponsored Recommendations


To join the conversation, and become an exclusive member of Electronic Design, create an account today!