Brain-like computer developed by Tsinghua University: “Completeness of brain-like computing” is proposed for the first time
chrispeng According to CCTV news reports, on the evening of October 14, Zhang Youhui’s team from the Department of Computer Science of Tsinghua University and Shi Luping’s team from the Department of Precision Instruments and their collaborators published a paper in the journal Nature, proposing for the first time “brain-like computing completeness” and software and hardware decoupling Brain-like computing system hierarchy. Currently, Tsinghua University is developing a new type of brain-like computer.
In this paper titled “A Brain-inspired Computing System Hierarchy”, the “brain-inspired computing completeness” and a brain-inspired computing system hierarchy with decoupling of software and hardware are proposed for the first time. The hardware completeness and compiling feasibility of the system expand the application scope of the brain-like computing system so that it can support general-purpose computing.
Brain-inspired computing is in its infancy and has not yet formed internationally recognized technical standards and solutions. This achievement fills the gap and is conducive to autonomously mastering the core technologies of new computer systems. The reviewers of “Nature” think “This is a novel idea and may prove to be a major development in the field of neuromorphic computing and artificial intelligence research.
Zhang Youhui, a researcher at the Department of Computer Science and Technology of Tsinghua University, believes that the concept of brain computing completeness is equivalent to helping us define what a brain-like computing system can do and where its functional boundaries lie. Then the hierarchical structure of the brain-like computing system can help us to realize it. For example, when we develop brain-like computing in the future, software developers, researchers, hardware developers, and researchers can do their own professional work relatively independently. Work.
Brain-like computing is a general term for computing theory, architecture, chip design, and application models and algorithms that draw on the information processing mode and structure of the biological nervous system. Its importance is as pointed out by the EU Human Brain Flagship Research Programme project: “Whoever wants to lead the world economy in the next 10 to 20 years must lead in this field.”
It is worth mentioning that on September 1 this year, Zhejiang University and Zhijiang Laboratory also released a brain-like computer, Darwin Mouse, in Hangzhou, which contains 120 million spiking neurons and nearly 100 billion synapses. This is my country’s first brain-like computer based on brain-like chips with independent intellectual property rights. The computer uses 792 “Darwin II” brain-like chips developed by Zhejiang University, supporting 120 million spiking neurons, nearly 100 billion For synapses, the number of neurons is equivalent to that of the mouse brain, and the typical operating power consumption is only 350-500 watts. It is worth mentioning that Darwin Mouse is also the largest brain-like computer with neurons in the world.
It can be seen in the official WeChat account of Zhejiang University that multiple robots use the brain-like computer as the intelligent center, demonstrating the collaborative work in flood fighting and emergency rescue scenarios.
Darwin Mouse can also realize the function of listening to songs and recognizing songs. People only need to sing two houses in a song. The brain-like computer can realize the access to the memory information in the brain by simulating the memory mechanism of the hippocampus. The content of the song “sings” out.
With the help of brain-like computers, the researchers also realized the real-time decoding of steady-state visual evoked potentials of EEG signals, which can be “thought” typing input.
The picture comes from the official WeChat account of Zhejiang University
The picture comes from the official WeChat account of Zhejiang University
The Links: LM64C15P SKD30-08A1
