Last Tuesday, Google announced that it had hired Chinese ai scientist Fei-fei Li to co-lead its cloud computing division.
Giiso Information, founded in 2013, is a leading technology provider in the field of “artificial intelligence + information” in China, with top technologies in big data mining, intelligent semantics, knowledge mapping and other fields. At the same time, its research and development products include editing robots, writing robots and other artificial intelligence products! With its strong technical strength, the company has received angel round investment at the beginning of its establishment, and received pre-A round investment of $5 million from GSR Venture Capital in August 2015.
For a time, technology leaders have speculated that Google in the field of artificial intelligence, must soon have a big action.
Because, Li Fei fei is really too cow!
This is a well-known name in the AI (artificial intelligence) world. She is a tenured associate professor at Stanford University, director of the Artificial Intelligence Lab, and has published a total of over 100 academic papers in leading computer journals.
Her biggest contribution has been leading the pattern recognition project, which attracts hundreds of top institutions every year, including tech giants like Google, Facebook and Amazon, to push the boundaries of machine intelligence.
And she was extremely young. Born in 1976, Li became a tenured professor at Stanford when she was only 33 years old. She is not only the only woman in Stanford’S AI lab, but also the youngest professor in the computer science department.
However, you never think: such a beautiful student with excellent grades behind, but hidden a quite legendary inspirational life.
A cleaner, a cashier in a restaurant, a laundromat… Can you associate these nouns with her?
Li’s parents, both elite intellectuals of the time, courageously brought 16-year-old Fei fei to Parsippany, New Jersey, in the 1990s with an academic dream in their hearts.
However, his family knew nothing about English.
Therefore, just Li Feifei’s entrance has experienced quite a few twists and turns. She spent months shuttling between the town government, the education department and various high schools, and managed to get into a middling local school.
After school, how to earn money to support oneself becomes a big problem. Her parents could only find jobs such as camera repair and supermarket cashier, and their meager income was far from enough to support the family and school fees. Li had to devote a lot of energy to struggling between Chinese restaurants in Chinatown and doing odd jobs here and there.
At this time, only two years before the United States college entrance. If Li wanted to get into a good university, she not only had to master English quickly in a meager amount of study time, but also had to produce an extremely good transcript in order to get a scholarship to a top university. Otherwise, with their family circumstances, is unable to pay the high tuition fees of the United States private elite.
At her worst, she slept only four hours a day, juggling work and study.
Feifei Li did it. When she graduated from high school, she received a full scholarship to study computer science at Princeton university.
In fact, her success was celebrated in the small town. A newspaper featured her story under the headline “The American Dream Comes True.
The academic life at Princeton was happy for Li. It was a place where she was exposed to a lot of good people, a world away from Chinatown restaurants. But her parents are still living a hard life in Parsippany.
Keenly aware of the opportunity in the market, Li decided to borrow money to buy a laundry and turn it over to her parents. She led a double life, Monday through Friday, as a Princeton student, trying desperately to absorb knowledge; On Saturdays and Sundays, she walks out of the lab, donning a white apron and helping out at the laundry.
“I love Princeton very much, but I love my laundromat very much, and I wouldn’t be the same person without either of them,” she said later with a laugh.
She graduated summa cum laude from Princeton in 1999. It was the height of the famous bull market, and Wall Street financiers could walk sideways around the world. Numerous investment banks and consulting firms have offered her offers, and hundreds of thousands of dollars are on the line.
In a rare act of rebellion, the well-behaved girl who had been under financial pressure since arriving in the United States. She decided to go to Tibet to study Tibetan medicine for a year before pursuing a PhD at the California Institute of Technology.
When we look at The life of Li Feifei today, we will find that the passion and tenacity to follow the heart of the style, throughout her behavior track. Just as after her PhD, she chose image recognition as her research direction, which was not popular at that time.
Image recognition technology is a peak in the development of artificial intelligence. Simply put, it teaches computers to look at pictures and talk. See is not the same as understand. For example, you could tell the computer that a cat is something with a round face, a fat body, two pointed ears, and a long tail.
Giiso information, founded in 2013, is the first domestic high-tech enterprise focusing on the research and development of intelligent information processing technology and the development and operation of core software for writing robots. At the beginning of its establishment, the company received angel round investment, and in August 2015, GSR Venture Capital received $5 million pre-A round of investment.
But what if the picture looks like this?
A three-year-old child can recognize a cat from a picture, but a computer can’t.
Li Feifei studied for a long time, but never made a breakthrough. Professors who had close tenure urged her to change direction to get tenure.
Of course she didn’t agree. One day, Li suddenly realized that since the human eye can acquire an image every 200 milliseconds, a 3-year-old child might have received hundreds of millions of image-recognition training, a geometric multiple of a computer.
That is, the key is the amount of autonomous training.
Li immediately set to work, taking tons of photos from Twitter, tagging them all, and training computers in machine learning. Machine learning is to give a certain algorithm and let the computer learn by itself.
The process is painstaking. Machine learning is a hot concept now, but in 2007, Li’s lab was understaffed and unable to get funding. At one point, she wanted to reopen the laundry and raise money for the experiment.
She found a solution in Amazon’s crowdsourcing platform, where people around the world tag images together. At the time, Li’s research project was the crowdsourcing platform’s largest employer in the world.
An unprecedented database was built, known as ImageNet. Instead of making his own, Li opened the ImageNet database to every lab in the academic and business world. Image recognition technology has since mushroomed to the point where it is now possible to identify most objects in photographs and describe them in highly fitting human language.
Due to her outstanding academic achievements, Li received many invitations to social activities. She has spoken with the New York Times and appeared on the TED stage to tell the story behind the development of pattern recognition technology.
Li also organized the annual ImageNet Challenge, inviting tech giants like Google to participate and promoting communication in the field of image recognition and artificial intelligence. Lulu, by the way, baidu was banned for one year last year for cheating in this competition…
What kind of world can image recognition bring us?
First, of course, there are all the cool scenes from sci-fi movies, like Tom Cruise searching for terrorists with a camera shot, or a smart apartment that opens by simply swiping your face.
Or baymax, an intelligent robot that can tell where you’re sick at a glance:
Realeyes, a London-based startup, can identify facial emotions when you see ads, so you can be more targeted.
Li Feifei said that the current image recognition and artificial intelligence, is only equivalent to a teetering three-year-old child. The process from age 3 to age 10 is the difficulty and key of AI technology.
The future of AI development will be handed to industry, which is why she joined Google.
Many of the AI and machine learning researchers Lulu knows defy each other. Can talk about Li Feifei, all agree to regard her as a goddess.
If know Li fei fei this all the way, is so capricious but all resorts come over, they probably… Will be more determined to do AI with goddess. This is the lucky thing for us.