Thu. Apr 25th, 2019

Lightelligence releases prototype of its optical AI accelerator chip

Accelerator chips that use mild relatively than electrons to hold out computations promise to supercharge AI mannequin coaching and inference. In idea, they may course of algorithms on the velocity of sunshine — dramatically sooner than at this time’s speediest logic-gate circuits — however up to now, mild’s unpredictability has foiled most makes an attempt to emulate transistors optically.

Boston-based Lightelligence, although, claims it’s achieved a measure of success with its optical AI chip, which at this time debuts in prototype kind. It says that latency is improved as much as 10,000 instances in contrast with conventional hardware, and it estimates energy consumption at “orders of magnitude” decrease.

The expertise underpinning it has its origins in a 2017 paper coauthored by CEO Yichen Shen. Shen — then a Ph.D. scholar finding out photonic supplies at MIT underneath Marin Soljacic, a professor at MIT’s Division of Physics who runs the college’s photonics and trendy electro-magnetics group — printed analysis within the journal Nature Photonics describing a novel technique to carry out neural-network workloads utilizing optical interference.

Lightelligence was based months later, and Soljacic was one of many first to affix its board of administrators.

“A scholar like Yichen comes by means of a couple of times in a professor’s profession, even at MIT. Yichen employed his background in nanophotonics and machine studying to invent this subject of utilizing optics for AI. He’s an actual pioneer,” stated Soljacic.

The chip in query — which is concerning the measurement of a printed circuit board — packs photonic circuits just like optical fibers that transmit indicators. It requires solely restricted power, as a result of mild produces much less warmth than the electrical energy, and is much less prone to adjustments in ambient temperature, electromagnetic fields, and different noise. It’s designed to fit into current machines on the community edge, like on-premises servers, and can ultimately ship with a software program stack appropriate with algorithms in generally used frameworks like Google’s Tensorflow, Fb’s Caffe2 and Pytorch, and others.

Lightelligence has up to now demonstrated MNIST, a benchmark machine studying mannequin that makes use of laptop imaginative and prescient to acknowledge handwritten digits, on its accelerator. And it’s recorded matrix-vector multiplications and different linear operations — key parts of AI fashions — working roughly 100 instances sooner than the state-of-the-art digital chips.

“We’re more than happy to disclose our working optical chip AI computing system,” Shen stated. “Our prototype … is 100,000 instances sooner than the system demonstrated in our Nature Photonics paper and a fraction of the dimensions. The system is a real testomony to our group.”

Up to now, Lightelligence has raised $10.7 million in enterprise financing and has over 20 workers, together with a lot of trade veterans hailing from Columbia, Georgia Tech, Peking College, and UC Berkeley. Headlining the roster is Dr. Gilbert Hendry, who’s held varied roles at Google and Microsoft, and Maurice Steinman, a former AMD senior fellow.

Lightelligence stands largely alone within the optical AI accelerator area, but it surely competes with Lightmatter, which has raised double the quantity of funding ($22 million) for its personal chip. (Lightmatter’s CEO Nicholas Harris, curiously, was a coauthor on that Nature Photonics paper.)

Leave a Reply

Your email address will not be published. Required fields are marked *