Monday , 24 October 2016
Breaking News

Intel Buys Deep-Learning Startup Nervana for $408M for AI Smarts

A two-and-a-half-year-old startup developing a new type of chip designed specifically for deep learning is set to become part of Intel. California-based Nervana, a company founded in 2014 by a team of machine-learning experts, is being acquired by Intel for a reported $408 million.

The startup is scheduled to release its Nervana Engine, a deep-learning-focused, application-specific integrated circuit sometime next year. Nervana’s technology and expertise will help Intel expand its capabilities in artificial intelligence, according to Diane Bryant, executive vice president and general manager of Intel’s Data Center Group. The company will offer more details about its plans in those areas during next week’s Intel Developer Forum.

‘Deep Learning at Ludicrous Speed’

“With this acquisition, Intel is formally committing to pushing the forefront of AI technologies,” Nervana CEO and co-founder Naveen Rao wrote today on his company’s Web site. “With this deal, we can now shatter the old paradigm and move into a new regime of computing. We’ll look back in 10 years and see this time as the inflection point of when compute architectures became neural.”

Intel’s Bryant added, “We believe that bringing together the Intel engineers who create the Intel Xeon and Intel Xeon Phi processors with the talented Nervana Systems’ team, we will be able to advance the industry faster than would have otherwise been possible.”

Still in development, the Nervana Engine is being designed from the ground up with deep learning and artificial intelligence applications in mind, according to Carey Kloss, the company’s vice president of hardware. Many of those applications today are supported by graphics processing units (GPUs), but those chips feature control logic, caches and other elements that were “optimized for video games, not deep learning,” he in a post on the company’s Web site earlier this year.

“When designing the Nervana Engine, we threw out the GPU…