News classification
Contact us
- Add: No. 9, North Fourth Ring Road, Haidian District, Beijing. It mainly includes face recognition, living detection, ID card recognition, bank card recognition, business card recognition, license plate recognition, OCR recognition, and intelligent recognition technology.
- Tel: 13146317170 廖经理
- Fax:
- Email: 398017534@qq.com
We've explained how artificial intelligence (AI) predicts the future, and how it can change the workplace and even create jobs.
Although self-driving car and robot technology may occupy the headlines, but artificial intelligence, deep learning and similar techniques may make their biggest impact almost become invisible, method is through a myriad of ways to simplify and accelerate the daily life and business.
With the nvidia corporate vice President Alex White (Alex White) as saying the deep learning herald a "new era of computing", it will open for all people and all things "strong and profound changes. The company is developing more sophisticated graphics processing units, in order to improve the ability of machine learning, support 1500 start-up companies at the same time, in order to understand where machine learning can be human belt.
Machine learning is not only used in cancer treatment, robotics and driverless cars, and other areas of the eye-catching, is being increasingly used in the daily business operations. Although the subject matter is very ordinary, it is very important indeed. Deep learning may completely change people daily work in data processing, makes them more creative, more a sense of accomplishment and eventually make all efforts to bear fruit. In other words, deep learning can do boring things, so you don't have to.
Is artificial intelligence the creator of jobs? For example, if you work in the fields of marketing and sales, deep learning can real-time tracking on all social media interaction between customers and your brand. It frees you from the hassle of data collation to pursue a more valuable goal to build a larger work advantage. In this case, deep learning can be a means of improving employment. As businesses improve, they grow and create more jobs for our humble people.
At least, that's the theory. We have every reason to doubt it, but it is quite logical. Last year, for example, e-commerce sales reached $2 trillion, and the figure is expected to double by 2020. Sales and marketing are an important part of the job market, which can be perfected through deep learning infrastructure.
White recently spoke at the deep learning society in London. Another outstanding speaker is Dr Anthony Morse (Dr Anthony Morse), he is a scholar, in his words, "to let the machine like a child to learn". In a video, Dr. Morse introduced an object to a humanoid robot, an orange star. Then he showed the star with a red ball, before the robot knew the red ball.
When asked to pick up the red ball, robot can infer from the previous instructions that two objects which is orange in the stars, then deduce which object is the red ball (this is the only thing that's left of table object). This relatively simple task to reveal a subtle and complicated world, the significance of this experiment is to: a child may be is the kind of person robot to have the ability to teach their knowledge.
Now that technology has taken off, the potential is huge. So how do these possibilities translate into the real life of ordinary people? Morse said the visual image recognition is often considered to be examples of machine learning, because it is easy to understand by all people, not to mention the depth study on visual identification is easy to go beyond the fact that human. Anyone familiar with Facebook automatic marking tools will recognize that it is useful to users and advertisers, though deep learning ability more pictures for social networks.
Morse argues that the ability to learn depth is broad and twofold: first, to improve existing automation; Second, can produce a new production line and, often, he says, for the former type of product, the development of factory production lines could turn into more complex systems, if they are replaced by deep learning, update can produce more advanced products, make the new products and markets become more attractive. From speech recognition and real-time translation to driverless cars, from the early cancer detection to the Morse so-called magic software (take a picture of someone's clothes immediately purchased online), there's something for everyone needs or weird feature.
Infinite possibilities? Indeed, it has far-reaching consequences. But is there an infinite possibility? That is not the case, according to Morse. "It can't solve every problem in the sun," he said. This sentiment is very different from the great potential for deep learning that people hear.
Dr Adam Grzywaczewski, who has similar views to Morse, attributes the recent development of "deep learning" to three big advances. First, the availability of big data has increased. People upload to YouTube 100 hours of video every minute, and upload 350 million photos a day to Facebook. Second, this scale of data has led to new deep learning techniques being studied and developed. Finally, the leap of computing power is a huge force for these technologies.
However, deep learning is limited in depth. The marked training data is often expensive or non-existent, moores said. In addition, you need to weigh the warning against the potential for deep learning. Understand how children learn and apply it on machine learning is no longer a science fiction film material, it is happening now, it is fast, and it happened on a commercial level.
Dr Grzywaczewski is keen to stress that deep learning ecosystems are dynamic and are changing. Nvidia corporate vice President alex white believes that related to the machine learning research mainly confined to academic field, and non-commercial fields, the wave will soon be washed to the shore.