The idea consists in making the Luminous chip compatible to popular software platforms on the basis of AI open source, such as TensorFlow and PyTorch.īy the beginning of June, 2019 in Luminous do not say when chips of the company go on sale. The company’s product is an AI training and inference chip that fits the computing power of supercomputers into a single chip, allowing businesses to avoid major bottlenecks that traditional processors face. MOUNTAIN VIEW, Calif.- ( BUSINESS WIRE )- Luminous Computing, the machine learning systems company building the world’s most powerful AI supercomputers, today announced the appointment of. Several other startups, such as Lightmatter and Lightelligence, try to use silicon photonics for acceleration of work of AI. Luminous Computing develops photonics chips to handle workloads required by the artificial intelligence industry. The approach used by Luminous relies on earlier work on neuronomorfny photonics which was written by the cofounder of the company Mitchell Nahmias, studying in Princeton University. The Mountain View, California-based start-up, founded in 2018 by Marcus Gomez, CEO, Mitchell Nahmias, CTO and Michael Gao (who has since left the company), plans to push revolutionary technological. Gomez told that the Luminous company intends to use silicon photonics which provides fast data transmission using light just as fiber optic cables work. In Tinder he among other things trained the AI models and came to a conclusion that such tasks should take away read minutes, but not days and hours. We use silicon photonics to break through these barriers, and were re-architecting the entire AI hardware stack in order to provide multiple-order-of-magnitude improvements in bandwidth and performance. The cofounder and the CEO of Luminous Marcus Gomez managed to work in due time both in Google, and in Tinder. Description Data movement bottlenecks are the single greatest limiting factor in developing novel hardware for machine learning. by Martin Giles JAdvances in computing, from speedier processors to cheaper data storage, helped ignite the new AI era. Gates invested in the developer of AI chips which use light for data transmission Luminous Computing has developed an optical microchip that runs AI modelsmuch faster than other semiconductors while using less power.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |