want
Quantum computing has entered a somewhat troubling time. While it is clear that quantum algorithms can be successfully run, the qubit counts and error rates of existing hardware mean that commercially useful problems cannot be solved at this time. So while many companies are interested in quantum computing and developing software for existing hardware (and paying for access to that hardware), their efforts are focused on preparation. doing. They want the expertise and abilities necessary to develop useful software once it’s ready to run on their computers.
For now, we’ll have to wait for the hardware companies to build a sufficiently robust machine (one that doesn’t have a definite delivery date at the moment). It may take years. It may be decades. Other than learning how to develop quantum computing software, nothing is obvious about the hardware for the time being.
But a company called QuEra may have found a way to do something less obvious. The technology the company is developing could eventually provide a path to quantum computing. Until then, however, it is possible to solve certain mathematical problems on the same hardware, and improvements in that hardware will benefit both types of computation. Also, in a new paper, the company’s researchers have expanded the types of computations the machine can perform.
maintain neutrality
QuEra’s qubits are based on Neutral Atoms, an established technology also used by at least one other quantum computing startup. Neutral atoms are typically used in general-purpose gate-based quantum computers, where computations can be performed through a series of logical operations performed on qubits. They could potentially do any computation, but there are certain computations that gate-based quantum computers can do that classical computers couldn’t.
In the gate-based mode of neutral atom quantum computers, nuclear spins are used as qubits. Atoms are moved and held in place by laser light, creating traps where the atoms are energetically favorable. By moving these traps, any two atoms can be placed next to each other to perform bond operations. The electron cloud usually prevents the nuclear spins from interacting with anything, resulting in a very stable qubit. However, after exciting the atom into the Rydberg state, the spin can be addressed. In the Rydberg state, one of its electrons is excited to a very high energy, creating a distant cloud with few atoms attached.
Neutral atoms therefore provide all the necessary tools for gate-based quantum computing. A long-lived quantum state, the ability to set and read that state, and the ability to arbitrarily connect any two qubits by placing them in close proximity. However, as with other gate-based quantum computers, the number of qubits and error rates are too high at the moment to prove anything more than empirical.
But there is another mode of operation that QuEra calls “analog mode”. This is based on a phenomenon called the Rydberg Blockade. This is a quantum phenomenon in which the presence of one atom in the Rydberg state reduces the probability of other nearby atoms being in the same state. By controlling the distance between atoms, you can effectively create a situation where only one set of atoms enters the state.
This allows two (or more) sets of atoms to be entangled in quantum superposition. Place the atoms at a distance where only one can enter the Rydberg state, and both can be exposed to enough light to excite the electrons. Only one of them can respond, and there is no way to predetermine which one will respond. Until we measure it, both atoms are equally likely to be in the Rydberg state and are in a superposition state. And, as in any entangled system, measuring one atom means the second atom must be in the opposite state.
constraints on constraints
Now imagine placing the third atom in line with the other two atoms. All atoms enter the superposition of states, but there are only two stable low-energy configurations due to Rydberg blockade. Either both edge atoms are in the Rydberg state or only the central atom is in that state. Add constraints to your system. Changing the geometry changes the constraints. If three atoms are arranged in a triangle with sides of equal length, there are three stable terminal states, each with one atom in the Rydberg state, all equally probable.
Adding more atoms imposes additional constraints on the stable final states of the system, and the exact nature of these states is geometry dependent. And the QuEra folks realized that a small cluster of atoms with one set of constraints can be bridged by atoms to additional clusters with completely different constraints. This leaves the final state set by the combination of the two constraints. This process can then be repeated until the geometry determines a large amount of constraints on the ground state of the system.
These constraints can represent a form of mathematical problem called maximum weight independent set. A geometry represents a desired set of properties, and the ground state it settles on represents a member of the set with a particular property. “We take advantage of that fact. [the atoms] QuEra’s Alex Keesling said: “And sometimes this is a grid, sometimes it’s a matter of graphs literally represented where the atoms are placed relative to each other.”
One of the key features of this type of problem is that as the size of the sets increases, it becomes increasingly difficult to find these maximal sets using conventional computers. Another is that it is what is called an NP-complete problem. This means that the solution can be obtained by transforming any other NP-complete problem to solve the maximum weight independent set problem. This means that QuEra’s machines can potentially solve a wide range of math problems when operated in this mode.
