According to the news from Tsinghua University on October 9th, Professor Wu Huaqiang and Associate Professor Gao Bin’s team from the School of Integrated Circuits at Tsinghua University have developed the world’s first fully integrated, memory-resistor-based computing chip that supports on-chip learning (machine learning directly on the hardware). They have made a significant breakthrough in the field of memory-resistor-based computing chips that support on-chip learning, which is expected to promote the development of artificial intelligence, autonomous driving, wearable devices, and other fields. The relevant achievements are published in the latest issue of the journal “Science.”
Memristor, as the fourth basic circuit element following resistors, capacitors, and inductors, can “remember” the charge that has passed through it even after power is turned off. It is considered a new type of nanoelectronic synaptic device. As early as 1946, the “father of computers,” John von Neumann, proposed and defined computer architecture using binary encoding, with separate storage and processing units for data storage and computation. However, with the increasing demand for data storage and computation in applications such as artificial intelligence, data transfer, and processing can be time-consuming, energy-consuming, and prone to “traffic congestion.”
In 2020, Qian He and Wu Huaqiang’s team built a complete memory-resistor-based computing system using a multi-array memory resistor. In this system, they efficiently ran convolutional neural network algorithms, successfully verifying image recognition capabilities. The energy efficiency was two orders of magnitude higher than that of graphics processing unit chips, significantly increasing computing power while achieving complex computations with lower power consumption and hardware costs.
For the same task, the energy consumption of this chip for on-chip learning is only 1/35 of that of dedicated integrated circuits (ASIC) systems under advanced processes, with the potential for a 75-fold energy efficiency improvement.
Recommended Reading:
- Intel vs Arm: The Future of PC Processors
- How Pigeon Learning Resembles Modern AI: Shocking Study
- GPU Tensor Cores: How They Boost AI and Graphics
Disclaimer: This article is created by the original author. The content of the article represents their personal opinions. Our reposting is for sharing and discussion purposes only and does not imply our endorsement or agreement. If you have any objections, please contact us through the provided channels.
