Top Ten Growth Technologies for Artificial Intelligence
In order to strengthen the forward-looking prediction of the new generation of artificial intelligence technology and grasp the global technological innovation trends and development trends, the China Electronics Society recently visited experts of artificial intelligence-related enterprises and colleges and universities, and selected ten of the most distinctive growth technologies. The main contents are as follows: - Antagonistic neural networks . A neural network architecture consisting of a constantly generating data neural network module and a neural network module that continuously determines whether the generated data is true or not, creating realistic original image, sound and text data, which is expected to greatly enhance machine translation and face recognition. The accuracy and accuracy of information retrieval will be applied to the fields of automatic driving and security monitoring in the future. - Capsule network . A multi-layered neuron module is constructed in a deep neural network to discover and store information such as the detailed spatial position and attitude of the object. This technology enables the machine to quickly identify the same object in different situations with less sample data, and has broad application prospects in the fields of face recognition, image recognition, and character recognition. - Cloud artificial intelligence . A technology that combines the working mode of cloud computing with artificial intelligence to centrally use and share machine learning tools in the cloud. It can effectively reduce the threshold for the use of artificial intelligence technology in terminal equipment, and is conducive to expanding the user community. It will be widely used in many industries such as medical, manufacturing, energy, and education in the future. - Deep intensive learning . Combine deep neural networks with reinforcement learning with decision-making ability to realize the integration of perception, decision-making or perception decision-making through end-to-end learning. This technology can significantly improve the efficiency of machine intelligence to adapt to complex environments, and has broad prospects in the fields of intelligent manufacturing, intelligent medical care, intelligent education, and intelligent driving. - Intelligent brain-computer interaction . Information interaction and functional integration between the nervous system and external devices is achieved by establishing a direct connection between the human brain and an external device with high biocompatibility. This technology makes human communication and communication more diversified and efficient, and will be widely used in clinical rehabilitation, autonomous driving, aerospace and other fields in the future. - Conversational artificial intelligence platform . It is a combination of speech recognition, semantic understanding, natural language processing, speech synthesis and other solutions to provide developers with an open platform with recognition, understanding and feedback capabilities. This technology can realize the natural interaction between machines and people in the dialogue service scene, and it is expected to be applied to smart wearable devices, smart homes, smart cars and other fields in the future. - Emotional intelligence . Artificial intelligence is used to simulate humanoid emotional responses such as expressions, moods, and emotions to create a virtual image with emotional attributes. It can give machines and devices a better ability to recognize, understand and guide human emotions, and bring more efficient and user-friendly interactive experiences to users. In the future, they will be frequently and deeply applied to intelligent robots, intelligent virtual assistants and other fields. - Neuromorphic calculations . That is to simulate the biological brain nervous system, on the chip to simulate the function of biological neurons, synapses and network organization, giving machine perception and learning ability. The technology aims to make the machine have the characteristics of low power consumption, high efficiency and high fault tolerance similar to the biological brain, and will be widely used in the fields of intelligent driving, intelligent security, and intelligent search. - Yuan learning . Combining neural networks with human attention mechanisms enables machine intelligence to have rapid self-learning capabilities. This technology enables machine intelligence to truly achieve autonomous programming, significantly improving the efficiency and accuracy of existing algorithm models, and will move artificial intelligence from a dedicated phase to a general phase in the future. - Quantum neural network . That is to use quantum devices to build neural networks to optimize the structure and performance of neural networks. This technology utilizes the characteristics of ultra-high-speed, super-parallel and exponential-level capacity of quantum computing to effectively shorten the training time of neural networks. In the future, face recognition and image recognition will be used. , and character recognition and other fields have been widely used. (Reporter Yu Sinan finishing) Jiangmen HomeMaster Electric Motors & Appliances Manufacturing Co., Ltd , https://www.homemasterjm.com