Artificial intelligence will grow more rapidly in 2016

Roman Janpoliski is director of the cybersecurity lab at the University of Louisville in the United States. He believes that in 2016, we will see the rapid development of convolutional neural networks in deep learning of computers, and the use of supercomputers will make this area the focus of artificial intelligence development in 2016.

Shallow learning is the first wave of machine learning, mainly because the computer system learns statistical laws from a large number of training samples and makes predictions on unknown events. In fact, this artificial neural network is only a shallow model.

Deep learning is a kind of unsupervised learning. It refers to learning more useful features by constructing multi-layer machine learning models and massive training data. The purpose is to build and simulate the neural network of the human brain and imitate the human brain to interpret data. Convolutional neural network is a kind of computer deep learning structure, and it is a research hotspot in the field of speech analysis and image recognition.

2015 is the year of artificial intelligence: Robots take on factory production, driverless cars become a reality, WiFi-connected Barbie dolls chat with children through voice recognition... In 2016, what will happen in the field of artificial intelligence? Experts believe that artificial intelligence in 2016 will be more rapid development with “stimulant”.

Robots are also currently used mainly in precisely set working environments, and they are highly efficient in precise, repetitive work. When encountering new missions that have never been performed and new environments without certainty, robots are often "stupid."

Andrew Moore is dean of the School of Computing at Carnegie Mellon University, a leader in artificial intelligence. He said that the National Academy of Sciences has convened technical experts, economists, and sociologists to study artificial intelligence instead of artificial problems. Artificial intelligence does not replace the production of blue-collar workers, but traditionally believes that they cannot replace and require people. White-collar workers interact with people.

Moore believes that artificial intelligence technology "feel" human emotions is the most important and most advanced direction in this research field. Jan Polisky believes that the ability of computers to understand language will eventually develop into a "seamless communication" between people and computers.

More and more accurate images, sounds, and facial recognition systems allow computers to better explore the emotional state of people. The development of this technology has broad application prospects in education, depression treatment, clinical prognosis assessment, smart customer service, online shopping and other fields.

In fact, some businesses have begun to use artificial intelligence technology to determine whether customers are happy or satisfied when shopping online. Some apparel companies use artificial intelligence to help customers find the most desirable products when shopping online. For example, a customer looks at a piece of clothing and says, "I want this style of jacket, but it needs to be warmer," and artificial intelligence customer service needs to be understandable. Customer requirements.

A number of experts believe that while research on artificial intelligence technology is making rapid progress, relevant ethical research should go ahead. More and more technology developers, computer scientists, and engineers feel that they must resort to the opinions of third parties to create programs when they are developing.

Take a driverless car as an example. When an animal crosses the road while driving, what should the system do? How to write code? How can the value of human and animal life be measured? Is one person more valuable than one hundred cats? One million?

Andrew Moore said that this is not just a problem that can be solved by scientists and engineers, someone must give an answer first.