- Tram Ho
After more than a decade of working in the US, Zheng Yefeng felt like he had encountered an invisible barrier. He also sees the gap in the field of artificial intelligence between China and the US is gradually narrowing.
Last year, Zheng – a researcher at Siemens Healthcare in New Jersey – made a decision that could solve both problems at once. He accepted an invitation to lead a medical research and development group at Tencent’s YouTu artificial intelligence lab in Shenzhen, which is dubbed China’s Silicon Valley.
” There are almost no opportunities for promotion if I am in the US,” he said, describing a dilemma facing the Chinese workforce working in US technology.
Because of the U.S.-China war, Chinese nationals in the US technology industry are also more strictly controlled, and people like Zheng choose to return home to work in the field of AI. every day, especially after the Chinese government sees AI as a national priority. The diverse applications of this technology have attracted billions of dollars in venture capital, created startups of “massive” value like SenseTime and ByteDance, and sparked a battle to attract talent among company.
This has created a strange symbiotic relationship between the two countries that are competing to become AI superpowers. The US side, with its preeminent higher education system, is the place to train and train Chinese AI experts such as Zheng (he earned a PhD at the University of Maryland after completing all bachelor’s and master’s degrees at Tsinghua University in China).
” Many professors in China have excellent academic ability, but in terms of the number of leading professors, the US is in the lead” – Luo Guojie, who accepted the invitation of Peking University to do Assistant professor after studying computer science in the US.
Among international students majoring in computer science and mathematics at US universities, the number of Chinese nationals has the third largest number after India and Nepal, calculated in the 2018-2019 school year, i.e. accounting for 19.9%.
When interviewed by the South China Morning Post, many Chinese AI engineers who previously decided to stay and work in the US after finishing their studies have shared some noteworthy information that we will mention in Sequel below. These students only agree to reveal their names due to the sensitivity of the issues they speak out.
A 25-year-old Beijing man, Lin, graduated from one of China’s best engineering schools before he went to the United States to study a master’s degree in computer science in 2017. Like some of his classmates , he realized that the teaching method in China was outdated.
” It is unimaginable that a final programming test still requires you to handwrite command lines, instead of running and trying on a computer,” Lin said. He currently works as a software engineer for Google in Silicon Valley.
” Although I still have to handwrite tests (in the US), we have more opportunities to practice in the lab, and be able to do our own projects,” he added.
A Facebook software engineer, Zhuang, also had a similar experience at the university he studied in Shanghai.
” Many engineering students (in China) still read outdated textbooks and lack practice in the lab,” – he said. ” AI programming techniques have advanced very rapidly over the past few decades, which means that many Chinese students do not have access to the latest updated knowledge in this field, at least within the classroom.”
Zhuang also emphasized that many classes in China are taught in Chinese, which means that graduates of English proficient engineering are traditional languages of the global AI research community. .
The United States is home to five of the world’s top 10 universities in the field of AI, including computer vision and machine learning, while China has only three. Carnegie Mellon University (CMU) in Pennsylvania ranked 1, while China’s Tshinghua University ranked 2 – according to CSrankings.
Owning leading research institutes and an open culture that encourages free speech, including the free access to the Internet, the US has become a “magnet” to attract the brightest AI students. In the world.
In 2018, 62.8% of PhDs in computer science, information technology, and computer engineering in the United States were awarded to students from abroad.
” Building the best universities is not easy,” said Gunther Marten, a senior EU delegation official who came to China on the sidelines of the World Internet Conference in Wuhan in October. ” University is a space of free speech, while in China it is not.”
The chart shows the number of foreign AI scholars living in the United States presenting a presentation at the Workshop on Neural Information Processing System 2018
When students studying AI in the US complete their research, most will take advantage of a law that allows them to stay in the country for the next 3 years to hone their work experience.
Among foreign scholars participating in the Conference on Neural Information Processing System (NIPS) in 2018 – a major machine learning event for AI professors, 87% of the scholars presented the presentation. are those who came to work at US universities or research institutes after completing their PhD program.
” China has many great universities and companies, especially in certain sub-areas of AI such as computer vision, but many people still hesitate to come to China because of its political environment, quality of life. , and workplace issues “- Remco Zwetsloot, a researcher at Georgetown University’s Emerging Security and Technology Center said.
Some Chinese AI engineers trained in the US say they are afraid of China’s 996 working culture: working from 9am to 9pm, 6 days / week. Chinese technology companies often require employees to work long hours to demonstrate their dedication.
Lin, a Beijing man who now works at Google, once practiced at one of China’s largest Internet companies, said: ” I work continuously from waking up to going to bed. At Google, I see difficult to understand because many people here only work until 5 pm, but Google is still the leading company in the world. “ Lin said he would be happy to return to China if the above 996 working culture becomes more comfortable.
Tshinghua University students in Beijing during the graduation ceremony
Chen, a graduate student at Carnegie Melon, recently accepted a job offer from Google, each of her internships at the unicorn AI headquarters in Beijing, SenseTime, where she had to work from 10am to 8pm. – 10 pm most days of the week.
A SenseTime spokesman said the company has been working flexible hours on all its employees.
Besides balancing work-life, Chinese graduate students look for jobs in Silicon Valley because of higher pay.
” If we pre-tax income, many of us are paid more than 1 million RMB (US $ 142,000) per year, but in China, the highest salary paid to newly graduated students is only 200,000 – 300,000. Yuan (28,000 – 43,000 USD) “- Chen said.
However, for Chinese people who plan to stay in the US for a long time, the big barrier they face is how to get a work visa, especially in the current war situation. Most of the employees in the field of AI are holding H-1B visas, which allow US companies to recruit foreign employees with professional qualifications such as IT, finance, and engineers.
However, the number of non-immigrant H-1B visas has begun to decline since 2016 when it peaked at 180,000 – according to the US Department of State, and U.S. technology companies have complained. bothering that a change in the Trump administration’s policy has made the visa processing process longer and more complicated.
In 2017, President Donald Trump asked for a review of the H-1B visa program, saying that he did not want American companies to hire cheap foreign workers, causing Americans to lose their jobs. He also wants to prioritize highly skilled people and limit the number of people who want to come to the US because there are family members here.
Science graduates from foreign countries can stay in the US on a student visa for up to 3 years, and during this time they must find a way to get a work visa, which is issued on a regular basis. invention is not disclosed. Foreign students who have been working in the US can apply for a green card, which allows them to become permanent citizens here.
After working for a major US technology company for nearly three years on a student visa, a Chinese software engineer said she was transferred to the US company’s Beijing office last year because H-1B work visa is not earned.
” Although there may be some unique circumstances, it seems that the current tense situation – at least in the past few months – has not led to noticeable changes in the total number of Chinese students who stay in the US after. graduate “- Zwetsloot said.
Some Chinese AI scientists use Twitter to announce their stay decisions. Chen Tianqi, who recently received a PhD at Washington University in Seattle, and Jun-Yan Zhu, a CMU and UC Berkeley alumnus, currently working for Adobe, each tweeted that they would work at Cernegie Mellon as the assistant professor next year.
To achieve the goal of turning China into the ” world’s main AI innovation center” by 2030, the Chinese central government has stepped up efforts to attract US-trained talents.
The increase in the number of foreign students studying for a PhD in computer science, information science, and computer engineering in the US
The Thousand Talents plan has brought back more than 6,000 Chinese students and academics abroad since it was announced in 2008, but because of tensions escalating with the United States, China had to suspend this initiative.
In the long run, China’s willingness to invest significant amounts of money in AI can bring home more Chinese talent for better recruiting opportunities. From 2013 to the first quarter of 2018, China attracted 60% of global investment in AI.
The Chinese government is investing heavily in this area, in which the city of Shanghai is setting up an AI fund worth 10 billion Yuan (US $ 142 million) in August, and the Beijing city government announced back in April that it would provide 340 million yuan ($ 48 million) in funding to the Beijing Artificial Intelligence Academy.
” More and more professionals like me are returning, and some have opened their own businesses,” said Zheng, a researcher at Siemens Healthcare who has just joined Tencent. ” Chinese people are more likely to find venture capital investments in China than in other countries.”
Source : Genk