Data capability to accelerate machine learning

December 06, 2017 09:40
Data capability has enabled artificial intelligence systems to enhance their self-learning ability. Photo: Thomson Reuters

The next big trend was a key topic in the fourth World Internet Conference held in Wuzhen recently.

During the event, Alibaba founder Jack Ma said humans "have explored less than 10 percent of our brains, and the machine created by that 10 percent knowledge won’t beat humans”.

Tencent founder Ma Huateng noted that internet plus advanced manufacturing will create smart manufacturing that combines software, hardware and services.

Meanwhile, Baidu founder Li Yanhong said that the population dividend for the internet industry is ending, and artificial intelligence will be the new growth engine.

We are in an era of information explosion thanks to mobile internet and different applications. Enormous amount of data is being created every day.

Technology has not only changed the way information is being generated, but also the costs of information and the speed of information dissemination.

We also quickly realized that there is huge value embedded in these abundant data.

The abundance of data gives rise to data capability, enabling artificial intelligence systems to learn directly from data.

For instance, to filter out scam emails, we can first build up a large sample database. A deep learning system can study the database and keep adjusting the criteria to improve the accuracy of detecting scam emails.

The combined power of big data and machine learning can do a lot for us, like the provision of guidance for career planning and suggestion of the most optimum travel route. Such technology can also help with health monitoring and disease preventions.

This is already happening, as shown by examples like self-driving cars, smart investment advisor, smart customer service and robot surgeons.

This article appeared in the Hong Kong Economic Journal on Dec. 5

Translation by Julie Zhu

[Chinese version 中文版]

– Contact us at [email protected]


Venture Partner of Sequoia Capital China, former head of the data committee and vice president at Alibaba Group.