ALIBABA LAYS OUT R&D ROADMAP FOR FRONTIER TECHNOLOGIES
CHRISTINE CHOU | SEPTEMBER 19, 2018
Alibaba Chief Technology Officer Jeff Zhang on Wednesday laid out an ambitious roadmap for research and development in several areas of emerging computing technology.
Speaking at the group’s Cloud Computing Conference 2018 in Hangzhou, he said that over the next three to five years,
Alibaba DAMO Academy will focus its efforts on developing quantum processors. In the meantime, it will keep driving
quantum development ahead, including building cloud-accessed, quantum-classical heterogenous systems for delivering quantum computing power as a utility and searching for super-fast quantum-classical hybrid algorithms to solve fundamental problems in machine learning, optimization and physics simulations.
“We have to leverage Alibaba’s advantage, not only to develop chips, but also the system and environments to test and apply it, using Alibaba’s platforms, artificial intelligence and computing power, to develop [quantum technologies] to their full potential,” said Zhang.
At the same time, DAMO will nurture and expand its partnership network to explore quantum-enhanced solutions for industries ranging from e-commerce, logistics, finance, materials and pharmaceuticals. Based on technologies developed, commercial products will gradually be introduced, bringing value to Alibaba customers.
DAMO, which was established last year, now has over 300 researchers in eight cities around the world. Its focus has been on machine intelligence, robotics, fintech, data computing and quantum computing, developing a simulator that mimics quantum circuits.
The academy expects to introduce its first artificial intelligence inference chip, called “AliNPU” mid-next year. The chip could potentially be used for autonomous vehicles, smart cities and smart logistics.
“We want to strengthen our data-processing capabilities. By the end of the year, we’ll be able to see Alibaba’s first neural network chip,” said Zhang, referring to AliNPU. “In our current tests, it is able to increase our image processing capabilities by fourfold.”
And DAMO, with an eye toward providing computational power for the group’s data centers and advancing its Internet of Things businesses, ranging from smart homes to smart logistics, is conducting more R&D in AI training chips and specialized AI chips.
Also during the conference’s busy first day, Alibaba said it was establishing a chip company, a subsidiary under the Alibaba Group that will focus on customized AI chips and embedded processors to further support Alibaba’s growing cloud and Internet of Things businesses, as well as to provide intelligent solutions for different industries.
And Alibaba announced a global, university-level math competition, aiming to discover and nurture the next generation of math geniuses and raise awareness of a field that underpins numerous fundamental sciences.
Mathematicians, including Yurii Nesterov of Catholic University of Louvain, Belgium, Frank Kelly at the University of Cambridge, Gang Tian of Peking University, Jianshu Li at The Hong Kong University of Science and Technology and Yitang Zhang at the University of California, Santa Barbara will mentor select students from the competition. In addition, the Academy offered nine excellent young science researchers a “2018 DAMO Academy Young Fellow” award during the conference.
Alibaba Lays Out R&D Roadmap for Frontier Technologies | Alizila.com