Trained artificial intelligence systems can now surpass humans in specific areas, and the rapid development of related software technologies is closely related to the advancement of dedicated chips. After the chip's support for artificial intelligence is more perfect, the Internet of Things (IoT) will be expected to evolve into AIOT (AI+IoT). The blooming of intelligent robots may still be just the beginning. The edge computing led by artificial intelligence terminal chips will bring more business opportunities.
Hou Junyuan, IIK project manager of the ITRI, said that artificial intelligence will bring about changes in all walks of life, thus changing the future direction. The hardware architecture of traditional artificial intelligence computing mainly includes a central processing unit (CPU), a graphics processing unit (GPU), and a field programmable array (FPGA).
In the past, the cost of artificial intelligence was very expensive, because the algorithm it executed was to simulate human brain computing, that is, huge parallel computing, which is completely different from the computing mode of existing computers. At present, there are many special artificial intelligence chips on the market, such as Intel's Intel XeonPhi and NVIDIA's artificial intelligence super chip Tesler P100 GPU, and the price is declining.
Hou Junyuan pointed out that the special artificial intelligence system in a specific field, due to the clear application background requirements, profound domain knowledge, and model calculation calculation is simple and feasible, the intelligence level of single test can now surpass human intelligence and achieve concrete results in many fields. The technical challenge today is how to develop low-power, high-accuracy cognitive computing, including new computing architecture circuit design, algorithms, and so on. In the future, artificial intelligence chips will accelerate various neural network algorithms including ConvoluTIonNeutralNetwork and RecursiveNeutralNetwork by specific algorithm accelerators. The biggest advantage of the dedicated chip is that its cost and power consumption are reduced, which will greatly improve the operating efficiency of the artificial intelligence algorithm.
Although the current mainstream research in the field of artificial intelligence is artificial intelligence computing on the server, there are more and more applications that need to perform real-time operations on the terminal. This computing architecture is called edge computing. This trend will change the design and technical requirements of the overall artificial intelligence computing system architecture.
In addition, with the introduction of artificial intelligence technology, IoT terminals have also been upgraded to various AIOT smart machines. Among them, the service robot will be the first to become the new battlefield of AIOT. IFR estimates that global service robots will have an output value of approximately $19.6 billion in 2015-2018, while home service robots account for 74% and entertainment and leisure robots account for 26%. 2017 will be a year of competition for these two robots. Home service robots such as Amazon Echo, ASUS COCOROBO, and American Jibo and Buddy in France will play the role of smart home hub.
resin encapsulated transformer,EI54 120v to 12v transformer,120v to 24v transformer,480v to 24v transformer
IHUA INDUSTRIES CO.,LTD. , https://www.ihuagroup.com