[Exciting] Chinese Academy of Sciences chip breakthrough, China's AI will counterattack

9ce78c23cae4eef33ad66a3502ac99f7.jpeg

   Recently, the Chinese Academy of Sciences has made a major breakthrough in the field of artificial intelligence chips. The new neural network accelerator chip jointly developed by the Institute of Computing Technology of the Chinese Academy of Sciences and the 53rd Research Institute of China Electronics Technology Group Corporation has successfully achieved the characteristics of high efficiency and low power consumption, and achieved up to 1000 times the calculation in typical artificial intelligence tests. Efficiency improvement. This chip uses a structure called "heart neural network", which is similar to the human heart nervous system, contains multiple processors and memory units, and realizes the coordination and coordination between different processors in a way similar to "heartbeat". data flow. Such an architecture can achieve efficient parallel computing, thereby improving computing efficiency. In addition, when designing this chip, the researchers also used some other optimization techniques, such as fast binary matrix multiplication and sparse matrix optimization, to improve computational efficiency and reduce power consumption while ensuring computational accuracy. This chip breakthrough will help promote China's independent innovation capabilities in the field of artificial intelligence chips, improve the competitiveness of China's artificial intelligence industry, accelerate the popularization and promotion of artificial intelligence technology, and provide new ideas and ideas for the development of artificial intelligence technology in the world. new plan.

d912b539236e6089d2b25a3a623a10e0.png

What is the future prospect of this chip?

The advantages of this chip are mainly reflected in high computing efficiency and low power consumption. It uses a variety of optimization techniques, such as fast binary matrix multiplication and sparse matrix optimization, etc., thereby improving computing efficiency while ensuring computing accuracy and reducing power consumption. In addition, the chip's neural network accelerator architecture is also very unique, enabling efficient parallel computing.

In the future, this chip will play an important role in many fields. For example, it can be used in artificial intelligence applications such as natural language processing, speech recognition, image recognition, machine translation, etc. In these fields, a large amount of calculation and data processing are required. The use of this chip can greatly improve the calculation efficiency and accuracy, and help people better realize the application of artificial intelligence technology.

In addition, this chip can also be used in other fields, such as medical care, transportation, and smart manufacturing. In the medical field, it can be used for medical image processing and disease prediction; in the transportation field, it can be used for traffic flow prediction and intelligent driving; in the intelligent manufacturing field, it can be used for robot control and intelligent monitoring.

Overall, this chip has broad application prospects and is expected to play an important role in the future artificial intelligence revolution.

Data Sources:

The Chinese Academy of Sciences announced: a new breakthrough in the field of optical computing chips! 【Analysis of Optical Chip Industry】https://t.qianzhan.com/caijing/detail/230614-6a95d3ab.html

1.5 to 10 times faster than Nvidia! Can domestic optical chips break the monopoly? http://www.myzaker.com/article/648a6c5d8e9f090cdf606af8

The Chinese Academy of Sciences announced a major breakthrough in optical computing, and the computing speed may exceed NVIDIA's broad application prospects丨Industry outlets

https://www.yicai.com/news/101784061.html

Detailed video:

https://ycalvod.yicai.com/vms-new/2023/06/rb/962edb0aa8a05dc3cc3668eb3f76ead5.mp4?auth_key=1686971067-0-0-36195d685585466a82ed340ae9173a88&ycfrom=yicaiwww

Guess you like

Origin blog.csdn.net/qq_16498553/article/details/131255700