Google today announced the launch of a universal quantum computer, Bristlecone, of 72 qubits, which has achieved a low error rate of 1% and is equal to the quantum computer of 9 qubits. Google believes that the use of Bristlecone can achieve quantum hegemony. Last week, IBM exposed the internal structure of its 50 qubit quantum prototypes. Google's brilliant performance in quantum bits and error rates, in a moment, advanced the point of the 2018 quantum hegemony competition, followed by a milestone in the Microsoft rumours.
This article uses the address: http://www.21ic.com/news/rf/201803/753821.htm
Today, the Google quantum AI laboratory research scientist Julian Kelly in the Google Research official blog, introduces a peer reviewed, 72 bit of Google's latest qubit general computer.
"We just started testing," said John Martinis, a physicist at Google. "We are very optimistic from what we know now." If everything works well, Martinis says, quantum hegemony may come within a few months.
Kelly said that the goal of Google Google AI Lab (Google Quantum) is to build quantum computers that can be used to solve real-world problems. Google's strategy is to explore recent applications with systems compatible with universal error correction quantum computers. In order to enable a quantum processor to run an algorithm beyond the classical simulation range, it not only requires a large amount of qubits. The processor is also very important in the low error rate assurance of the readout (readout) and logic operations, such as the single and two bit gates.
Google presented a new quantum processor Bristlecone at the annual meeting of the American Physical Society in Losangeles. The gate based superconducting system aims to study the system error rate and extensibility of quantum bit technology, as well as the application in quantum simulation, optimization and machine learning.
A record 72 qubit quantum computer, with a 1% error rate, can achieve quantum hegemony.
Julian Kelly introduced the latest device to follow the physical principles of the linear array technology of 9 qubit quantum computers previously proposed by Google, which showed the best results as follows: low reading error rate (1%), single qubit gate (0.1%), and the most important double qubit gates (0.6%). The device uses the same mode as 9 qubits to be coupled, controlled and read out, but extends it into a square array of 72 qubits.
In the experiment, the researchers chose this size device to show the future quantum hegemony, using surface coding to study first and two order error correction, and to promote the development of quantum algorithms on the actual hardware.
1.jpg
Before studying specific applications, it is important to quantify quantum processors. The Google theory team has developed a benchmark tool to accomplish this task by using random quantum circuits on the device to randomly allocate the tasks of the system and to check the distribution of the sampling output by a classic simulation method. For a quantum processor with insufficient operating error, it can have a performance beyond the classic supercomputer on a clear computer science related problem, namely, "quantum hegemony". These random circuits must be large in qubits, computation length and depth.
Although no one is able to achieve this goal, Google researchers calculate that the target of quantum hegemony can be proved perfectly by using 49 qubits, a circuit depth of over 40, and a 2 bit error of less than 0.5%. They believe that this quantum processor is better than the supercomputer's experiments, and will prove to be a watershed in this field, and will also be one of the main goals of the future.
2.jpg
Google originally wanted to achieve similar performance on 9 qubit devices, but now it has achieved 72 qubits. They point out that Bristlecone will be a convincing proof of the future when building larger quantum computers. Google also said that running devices like Bristlecone under low system errors requires software, control electronics, and a variety of technologies, such as the processor itself, so it will need to be carefully observed in several rounds of iteration.
The Google quantum AI laboratory has pointed out that using Bristlecone can achieve quantum hegemony, and learning how to build and operate equipment at this level will be an exciting challenge. They expect to share the results, which can help more researchers to do experiments in this field.
Google: from 49 to 72, we are not only a little better than IBM.
As mentioned earlier, Google has said that it will build a quantum computer with 49 qubits to achieve quantum hegemony, which is the result of their calculations. Why did we jump to "72" qubits this time?
All along, we believe that the 50 qubits of quantum computers are the "starting price" of quantum hegemony. Shortly after Google's statement of quantum hegemony with 49 quantum bits of quantum computers, IBM said their research showed that for certain quantum applications, 56 and even more qubits could be needed to achieve quantum hegemony.
In addition, in November 12, 2017, IBM announced the successful development of a quantum computer with 20 quantum bits and successfully completed and tested the first 50 quantum bit computing prototypes of the world's first quantum bit.
IBM Inaugural Index, held in San Francisco last week
 

• 上一篇内容: 暂时还未添加
  
• 下一篇内容:暂时还未添加