customContents

entityId

entityUuid

articlePageQuery

filter

article

articleBaseAdvertisingInfoArticle

articleListArticle

articleSeoArticle

helpersCheckIsPostiesArticle

hooksArticleAcknowledgementGateOverlayArticle

hooksTrackPageViewArticle

hooksContentInterestArticle

articleLinkingDataContent

articleSchemaContent

liveArticle

advertZone

sentiment

contentLock

topics

types

sections

moreOnThisArticles

urlAlias

commentCount

sponsorType

authorLocations

authors

paywallTypes

corrections

displaySlideShow

multimediaEmbed

printHeadline

seriesContainer

socialHeadline

headline

flag

firstTopic

glossary

flattenTypeIds

publishedDate

additionalInfoBoxes

images

image

relatedLinks

relatedNewsletters

__typename

createdDate

sponsor

factSheets

series

knowledgeQuestion

disableOffPlatformContent

published

updatedDate

summary

subHeadline

body

keywords

copyrighted

customTimezone

liveContents

Zhang Tong

A team of scientists in Beijing has developed the world’s first fully optical artificial intelligence chip that they say significantly boosts efficiency and performance.
The Taichi-II chip was created by a Tsinghua University team led by professors Fang Lu and Dai Qionghai, whose study was published in the journal Nature on Wednesday.
It is a major upgrade from their earlier Taichi chip that the researchers in April said had exceeded the energy efficiency of Nvidia’s H100 GPU by over a thousand times.
Training artificial intelligence using the Taichi chip requires the assistance of electronic computers. But Taichi-II can be used for modelling and training based entirely on light – and that has made it much more efficient and improved performance, according to the team.
They said in the paper that the upgrade was a key step for optical computing and could help to take it from the theoretical stage to large-scale experimental applications, as well as address the growing demand for computational power with low energy consumption.
It could also provide an alternative after the US restricted China’s access to the most powerful graphics processing unit, or GPU, chips for AI training.


Taichi-II’s performance was found to be superior to its predecessor in various scenarios, according to the paper.
It sped up the training of optical networks with millions of parameters by an order of magnitude and increased the accuracy of classification tasks by 40 per cent.
In the field of complex scenario imaging, Taichi-II’s energy efficiency in low-light environments improved by six orders of magnitude.
Fang said conventional optical AI methods typically involved emulating electronic artificial neural networks on photonic architecture – based on light – designed on electronic computers.
“Because of system imperfections and the complexity of light-wave propagation, perfectly precise modelling of a general optical system is impossible, and a mismatch always occurs between the offline model and the real system,” she said.
To overcome these challenges, the team developed a method where a computer-intensive training process is conducted directly on the optical chip – so most of the machine learning can be carried out in parallel. They called this fully forward mode, or FFM, learning.
“This architecture enables high-precision training and supports large-scale network training,” wrote Xue Zhiwei, lead author of the study and a doctoral student.
FFM learning takes advantage of commercially available high-speed optical modulators and detectors, and it could do better than GPUs in the accelerated learning stakes.
“Our research envisions a future where these chips form the foundation of optical computing power for AI model construction,” Fang said.


Why death of star scientist Li Haizeng, 34, has Chinese researchers taking note

Starlink radiation makes stealth target glow on Chinese radar

China sets some global standards for 6G tech as it looks towards next-gen communications

LandSpace’s Zhuque-3 reusable rocket test a ‘breakthrough’ for China space race

Science

Chinese team creates world’s first AI training system that runs entirely on light

The scientists said their AI chip, known as Taichi-II, was a key step for optical computing. Photo: Shutterstock

Could China’s Taichi light-based chip show the way on AI?

Researchers are in race to develop energy-efficient chips for AI tasks. Photo: Shutterstock

Taichi could transform input images into works in the manner of various artists. Credit: Dai Qionghai

Tech war: China’s chip imports surge as firms stockpile ahead of fresh US restrictions

The surge in semiconductor imports reflects expectations that the US will impose fresh unilateral restrictions on mainland’s China’s access to high-bandwidth memory chips. Photo: Shutterstock

A 3D rendering of high-bandwidth memory chips bundled with a processor inside a server. Photo: Shutterstock

Mainland China’s semiconductor exports from January to July reached 166.6 billion units, up 10.3 per cent from a year earlier. Photo: Shutterstock

‘Godmother of AI’, shy Chinese-American pioneer Fei-Fei Li seeks science chance for all

The Post profiles Fei-Fei Li, the Chinese-American computer scientist who has become known as the “Godmother of AI”. Photo: SCMP composite/Huxiu/Sohu

Artificial intelligence pioneer Fei-Fei Li is fighting for diversity in the field she knows best. Photo: Huxiu/Sohu

Li has studied and worked at elite US institutions to earn the title “Godmother of AI”. Photo: AFP

Is this Chinese AI model a seismic change in earthquake prediction?

The AI model has been trained on seven years of earthquake data. Photo: Shutterstock

News

China

Taichi-II chip is fully optical and significantly boosts efficiency and performance, according to Tsinghua University scientists


Taichi-II chip is fully optical and significantly boosts efficiency and performance, according to Tsinghua University scientists.
