Why is It so Difficult for Traditional Industries to Get AI Blessings? — Part1

Whether it’s personalized recommendations for short videos, time-consuming estimates for takeaway delivery, or face recognition for mobile payments, AI technology is “handy” in the Internet industry. However, when it comes to traditional industries, it is difficult for people to quickly think of very mature typical cases of AI applications.

Why is the application speed and scope of AI technology in traditional industries far inferior to industries such as the Internet?

The application of AI in the consumer Internet industry has more advantages.“The application of AI technology mainly depends on data, computing power, and algorithms.” Zhu Pengfei, associate professor of the Department of Intelligence and Computing at Tianjin University, introduced that data must first reach a certain size, which is the basis of the application. In addition, computing power must be able to support large-scale Model training, and then the algorithm needs to achieve a certain accuracy. The end-to-side computing power must also have a certain reasoning ability.

Learn more: Data, Algorithms, and Processing are Three indispensable Elements of AI

The reason why only Internet companies are currently applying AI technology on a large scale is that Internet companies have more advantages in these three aspects.

In the past few years, short videos were not as popular as they are now. For example, Taobao, a giant Chinese e-commerce platform(which belongs to Alibaba group) at the early stage of its development did not have strong user stickiness. As the user recommendations become more and more accurate, the user experience has also been greatly improved. Therefore, AI technology ensures a blowout of user growth.

“Accurate push mainly relies on the improvement of algorithm accuracy, and the improvement of algorithm accuracy is inseparable from massive data.” Zhu Pengfei explained that in this scenario, the algorithm model needs to continuously learn and evolve. Since it is not a closed data pool and new data is always added, and the algorithm model needs to be continuously adjusted and iteratively upgraded through ML to make it more and more accurate, forming a virtuous circle.

“At the same time, although the accuracy of algorithms in the Internet industry has risen to a certain level, compared with the application scenarios of some traditional industries, the Internet industry has a relatively low acceptance threshold for the accuracy of AI algorithms. For example, Taobao’s preference recommendation and Baidu’s hot search keywords only need to achieve the purpose of user stickiness. As long as there is a certain accuracy, users can accept more easily.”

Zhu Pengfei said that in contrast, in many traditional industries, the requirements for technical precision are much higher. For example, in the application of vision-based AI technology (face recognition), identity verification at high-speed railway stations and airports, 1:1 comparison accuracy must be as high as 99.99% or even higher before it can be applied.

In terms of computing power, the current cloud computing power can already support large-scale model training and inference, such as Taobao’s recommendation system. However, in a large number of traditional industry application scenarios, the end-to-side computing power on smart terminals cannot meet the real-time requirements.

“Compared with social networks and e-commerce systems, the closed ecosystem of traditional industry application scenarios makes cloud computing power unable to be effectively used.” Zhu Pengfei said. Taking intelligent unmanned system inspections as an example, power inspections, pipeline inspections, traffic inspections, river inspections, and photovoltaic inspections all require the computing power carried on drones and robots to meet the requirements of real-time inspections.

However, due to the high complexity of the video analysis model, the end-side often fails to achieve accurate and efficient real-time Inference. As the accuracy of the algorithm can’t meet the requirements, the application of AI technology cannot be implemented in many scenarios.

Scalable and High quality labeled data is needed

With the acceleration of the commercialization of AI and the application of AI technologies such as assisted driving and customer service chatbot in all walks of life, the expectation of data quality in the special scenarios is getting higher and higher. High-quality labeled data would be one of the core competitiveness of AI companies.

If the general datasets used by the previous algorithm model are coarse grains, what the algorithm model needs at present is a customized nutritious meal. If companies want to further improve certain models’ commercialization, they must gradually move forward from the general dataset to create the unique one.

End

Outsource your data labeling tasks to ByteBridge, you can get the high-quality ML training datasets cheaper and faster!

• Free Trial Without Credit Card: you can get your sample result in a fast turnaround, check the output, and give feedback directly to our project manager.
• 100% Human Validated
• Transparent & Standard Pricing: clear pricing is available(labor cost included)

Why not have a try?

Source:https://baijiahao.baidu.com/sid=1713944602772076382&wfr=spider&for=pc