The quantum age is really coming!
Editor’s note: Google claims to have achieved “quantum hegemony”, IBM Microsoft is also developing quantum computers, it can be said that quantum computing has become the hottest cutting-edge technology! In the end, what is quantum computing, there are many introductions and explanations. Today, this article only has 2000 words, but it briefly introduces quantum computers, and students who want to learn from science can read them. This article is compiled from the original article titled “Quantum Computing: An Introduction for Programmers”.
A quantum computer is no better than a human being when solving complex problems. Quantum computers break down complex tasks into many simple tasks. Compared with humans, computers are much faster when dealing with simple tasks. This is the advantage of computers. But classic computers have limitations: tasks must appear in order. Because of this, if the task is too complicated, or the database is too large, it will take a long time to find a solution. Many times the problem is too large. From a mathematical perspective, even the most powerful supercomputer has no way to break through the obstacles of sequence task setting, but quantum computers can, because it has some interesting features: superposition, entanglement and interference.
How it works
To explain this phenomenon, we take a step back. What is the simplest task when a computer breaks down complex tasks into simple, small tasks? It is to choose between two options, such as A or B, true or false, head or tail, these are binary problems. In a computer, a binary code (represented by 1 or 0) can be converted to “on or off” in a computer circuit switch. Although binary solutions (information bits) can exchange information at an alarming rate, they must be read one by one when reading. Quantum computers are much more efficient. Equivalent to a bit is a qubit, which is essentially equivalent to a particle that can carry measurable information.
Bits must exist in a binary state or in another state, but qubits can exist in a quantum state (superimposed), which can exist in two states at the same time. Quantum mechanics is largely a probabilistic game. The probability that a qubit becomes state A or B may be 50/50, or 70/30, 10/90 or other ratios. You can imagine this: the position of the qubit is between AB, or at a certain position on the sphere. One end of the ball is in the A state and the other end is in the B state. No matter what, because the quantum has superpositionFeatures so it can appear in multiple locations at the same time. In order to find a solution to the problem, the qubit can advance along multiple paths at a time, but the bit can only select one at a time.
Dijkstra can help us find the most efficient path to the destination. Qubits don’t have to be explored one by one (the classic computer does just that), it can analyze multiple paths at the same time. Find the best path faster. The more complex the problem, the larger the input information, and the longer the classic computer will find the path. Quantum computing is different, and it is much more efficient.
To exploit the advantages of quantum superposition, time is critical because the superposition characteristics of qubits are affected when they are in contact with the measurement device. We call this physical law the “observer effect.” Although particles show the characteristics of particles and waves at the same time, only one of them can be recorded when we observe. Which one of them is recorded depends on the observation. So, when we want to know what kind of information a qubit carries, we face such an obstacle.
We can use the second feature of quantum mechanics to overcome the “observer effect,” which is characterized by “entanglement.” The physicist has confirmed the existence of “entanglement”, that is, the two particles can be linked regardless of how far apart. Now we can manipulate dozens of qubits and turn them into a single entangled state, so that we can build a network with 2 n-th power possibilities (n is the number of qubits in the network), they can Collaborative work.
What if the qubits carry the same information? That is to talk about quantum interference. Particles have the characteristics of waves, and interference is one of the characteristics of waves. When the crests meet the crests, the troughs meet the troughs and complement each other, and the effect is magnified. This is constructive interference. If the peak meets the trough, it will cancel out, which is the destructive interference. When more than one qubit is in constructive interference, their effect is amplified and the information can be transmitted.
Where is it now?One step
There are still some obstacles to be made to make quantum networks truly realize their potential. Although quantum computers solve problems faster than classical computers (the so-called quantum advantage), even today’s largest and most stable quantum systems have no practical value in business.
In fact, adding qubits to the entanglement system is very difficult because the network is very fragile. In 1998, IBM, Oxford, the University of California, Berkeley, Stanford, and MIT successfully combined a pair of qubits. Twenty years later, Google set a new record, increasing the number of qubits to 72.
Although entanglement can solve the problem of “observer effect” to a certain extent, the quantum state is still easily destroyed, and the duration of quantum features is also limited. The quantum system must find a solution before exiting the superimposed state and entering the decoherent state, otherwise it will fail.
External factors can also cause qubits to exit the superposition state. Although we can increase the number of qubits, the more qubits, the more susceptible it is to external factors. Nowadays, the industry generally creates an environment with lasers, magnetic fields, and superconductors to extend the life of the quantum state (the lifetime is usually calculated in milliseconds), which can reduce the “error rate.”
When the error rate drops, the observing system may make a breakthrough, and we can develop better quantum algorithms based on observations. Some industry players have allowed customers to enter the quantum computing network through the cloud, which makes R&D easier.
Once we can build a quantum bus that is large enough and stable enough, once the error rate drops low enough, the quantum computer will solve the classic problem faster, not only that, but it can also solve the problem that the classic computer can’t solve. .
At this stage, “quantum hegemony” can be achieved. Others believe that “quantum hegemony” is impossible to achieve, because it is limited by physical principles and theory, quantum computing can not go this step.
What is the possibility?
Once quantum quantum hegemony is truly realized, quantum computing can come in handy in many scientific fields to solve complex problems. When queried on a complex and large database, it processes faster; machine learning will advance by leaps and bounds; we can simulate more complex molecular structures and understand their behavior so that we can make more breakthroughs in the medical field.
With powerful simulation capabilities, it is also good for the industrial and technology industries. But quantum computers can’t replace classic computers, they need to be combined with modern machines. With quantum computers, some areas will usher in change.
When AI, machine learning and quantum computing are combined, there may be a big breakthrough. The cybersecurity industry will also embrace quantum technology, because even today’s best classic encryption technology is in the face of quantum systems.
Translator: Xiaobing Hand