According to the People’s Daily, the team of Shi Luping, a professor of precision instruments at Tsinghua University, recently released its research results, the brain-like computing chip “Tianjicâ€, the world’s first heterogeneous fusion brain chip, and the world’s first. An artificial intelligence chip that supports both pulsed neural networks and artificial neural networks.
Based on the research results, the paper "Towards artificial general intelligence with hybrid Tianjic chip architecture" was published as a cover article on August 1st, "Nature", which realized China's Zero breakthrough in the two major fields of chip and artificial intelligence.
Heterogeneous fusion of "sky movement"
In 2015, the first generation of "day movement" came out. The first generation of chips at that time was about 110 nanometers in size, just a DEMO (demo sample).
After continuous improvement of the design, the team developed the second generation of “Tian movement†chips in 2017.
The second generation "sky movement" is reduced in size to 28 nanometers, featuring high speed, high performance and low power consumption.
Compared to the current world-class IBM TrueNorth chip, it has more complete functions, flexibility and scalability, 20% higher density, at least 10 times faster, and at least 100 times faster bandwidth.
It is understood that the Celestial Chip has a number of highly reconfigurable functional cores that can support both machine learning algorithms and brain-like circuits, consisting of 156 FCores, containing approximately 40,000 neurons and 10 million synapses, using a 28nm process. Process, the area is 3.8 × 3.8mm?.
In general, because the language, calculation principle, coding method and application scenario of the two types of models are different, the computer science model and the neural network model are supported by the deep learning accelerator and the neuromorphic chip, respectively.
It is worth noting that as the world's first heterogeneous fusion brain chip, supporting both computer science models and neural network models is a major feature of the Celestial Chip.
Team from Tsinghua University
The author of the paper "Towards artificial general intelligence with hybrid Tianjic chip architecture" comes from Tsinghua University, Beijing Lingbi Technology, Beijing Normal University, Singapore Polytechnic University and the University of California, Santa Cruz. Barbara and other institutions.
Among them, Tsinghua University is the core. Shi Luping, professor of the Department of Precision Instruments at Tsinghua University and director of the Brain Computing Center, is the author of the paper. Six of the seven first authors are from Tsinghua University and the other is from the University of California, Santa Barbara.
Since joining the Singapore Academy of Sciences, Institute of Data Storage, Singapore Academy of Sciences, Schlupin participated in the creation and leadership of the Institute's research in semiconductor non-volatile memory, optical storage, and artificial cognitive memory.
In 2012, Shi Luping gave up Singapore's excellent treatment and accepted the invitation of Qiu Yong (currently the president of Tsinghua University), then the head of Tsinghua University, to join Tsinghua University to participate in the creation of a brain-like computing research center.
“This is a very promising field, but it is also very risky and challenging.†Shi Luping said that the team set a goal of developing brain-like computing and supporting artificial general intelligence.
At the press conference, Professor Shi Luping introduced the research ideas of the paper. He mentioned that there are two main methods for developing artificial general intelligence at this stage: one is based on computer thinking, and the other is based on human brain thinking. Although each has its own advantages and disadvantages, it all represents some models of human brain processing information. Their research team proposed an architecture that heterogeneously blends the two methods, and developed a platform chip on this architecture.
In addition, he also revealed that the team has started research on next-generation chips and expects to complete R&D work early next year.
Automatic bicycle demonstration
In order to prove the feasibility of building a brain-like paradigm intelligent system, Shi Luping led the team to use it to create a "brain" and install it on a bicycle.
Based on the “brainâ€, the bicycle achieves autonomous riding, enables real-time object detection, and enables autonomous obstacle avoidance, S-type tracking, and voice command recognition, acceleration, deceleration, steering, control balance, and decision making while tracking humans. Wait for the task.
To achieve these tasks, the “brain†needs to overcome three main challenges:
First, in the outdoor natural environment, it successfully detects and smoothly tracks moving targets, crosses the speed bumps, and automatically avoids obstacles when necessary.
Second, real-time motor control signals are generated in real-time response to balance control, voice commands, and visual perception to keep the bike moving in the right direction.
Third, to achieve integrated processing and rapid decision making of a variety of information.
To accomplish these tasks, the team developed several neural networks, including CNN for image processing and object detection, CANN for human target tracking, SNN for voice command recognition, and MLP for attitude balance and direction control. There is also a hybrid network for decision control.
Thanks to the chip's decentralized architecture and arbitrary routing topology, the Tianjic chip can simultaneously support computer science-based machine learning algorithms and neuroscience-based biological models, freely integrate various neural networks and hybrid coding schemes, and implement multiple models. Seamless communication between the bikes enables the bike to perform these tasks smoothly.
Universal Artificial Intelligence (AGI)
At this stage, there are two main methods for developing artificial general intelligence: one is based on neuroscience, and the human brain is simulated as much as possible; the other is computer science-oriented, allowing computers to run machine learning algorithms.
Both have their own advantages and disadvantages, and the integration of the two is currently recognized as one of the best solutions. Developing a computing platform that combines the two will be a key to driving convergence.
The new chip combines two technical routes, and this fusion technology is expected to enhance the capabilities of each system and promote the research and development of artificial general intelligence.
Named "Tianjic", this hybrid chip has multiple highly reconfigurable functional cores that support both machine learning algorithms and existing brain-like algorithms.
At present, the technology has a certain foundation. In addition to bicycles, the development of related technologies can be directly extended to applications such as autonomous driving, drones and intelligent robots. In the future, Professor Shi Luping will lead the team to closely link research and market.
Thoughts triggered by the development of science and technology
The development of science and technology has brought us great convenience. At the same time, it has also caused people to think, worry about the "terminator" in the movie, and whether the "Matrix" will become a reality. What kind of situation will people face at that time?
The "brain-computer interface" research conducted by Musk, recently called "Silicon Valley Iron Man", has caused widespread debate.
Mask's Neuralink presented its findings at the latest product launch in July: within ten seconds, the laser drilled holes in the skull, and the robotic “sewing machine†inserted half of the human hair's hair into the brain, precisely avoiding the blood vessels. The brain-to-machine information transfer begins.
A new type of chip has been successfully developed and has been successfully implanted into the brain of a mouse to control the behavior of the mouse. This makes many people feel new, but the industry has different views on this behavior.
Recently, Mei Hong, an academician of the Chinese Academy of Sciences, publicly stated that I strongly oppose Musk's research on the implantable brain-computer interface. This technology is currently very dangerous, although the future human brain mechanism is imperative for technological development. But before the technology matures, it should not be so bold and so radically researched. If technology is abused, the consequences are unimaginable.
Mei Hong’s words were loud and sound, and many people also sounded an alarm.
Can the brain like the brain?
As the world's first heterogeneous fusion brain chip, its powerful processing power makes us stunned, but also can not help but think, so such a smart "brain" will transcend humans?
In this regard, Shi Luping also expressed his opinion: the computer has long surpassed the human brain, just to say in what areas. However, at present, at many levels of intelligence, especially for uncertainties, there are still considerable distances between computers and human brains in many fields such as learning and self-determination.
It is undeniable that artificial intelligence is a very hot topic in recent years. It brings great convenience to human beings, and it also attracts people's extensive attention and some doubtful voices. However, as the creator of artificial intelligence, I believe that human beings can make good use of our wisdom to master it, control it, regulate its development path, and benefit it for our human beings to avoid those risks to the utmost extent.
Here you can find the related products in Fan Clutch, we are professional manufacturer of cooling fan clutch,electric fan clutch, ac fan clutch,engine fan clutch, fan clutch price , car fan cluth ect.
We focused on international export product development, production and
sales. We have improved quality control processes of Fan Clutch to
ensure each export qualified product.
If you want to know more about
the products in Fan Clutch, please click the product details to view
parameters, models, pictures, prices and other information about cooling fan clutch,electric fan clutch, ac fan clutch,engine fan clutch, fan clutch price , car fan cluth.
Whatever
you are a group or individual, we will do our best to provide you with
accurate and comprehensive message about Fan Clutch!
Fan Clutch Tool,Fan Clutch Assembly,Fan Clutch Autozone,Fan Clutch Adapter
Guangzhou Taixin Auto parts Trading Co., Ltd. , https://www.hyundai-autopart.com