Cuella
It is widely agreed that the most useful quantum computing will have to wait for the development of error-correcting qubits. Error correction involves distributing bits of quantum information, called logical qubits, among a small collection of hardware qubits. The disagreements mainly focus on the best way to implement it and how long it will take.
An important step toward that future is described in a paper published today in Nature. A large research team, primarily based at Harvard University, has demonstrated the ability to perform multiple operations on up to 48 logical qubits. This study shows that a system based on the hardware developed by QuEra can accurately identify the occurrence of errors, which significantly improves calculation results.
Yuval Boger, QuEra’s chief marketing officer, told Ars: “We feel this is a very important milestone on the path to what we all aspire to: large-scale, fault-tolerant quantum computers. I am.
Catching and fixing errors
Complex quantum algorithms can take hours to maintain and manipulate quantum information, and existing hardware qubits will likely never reach a level where they can process quantum information without making errors. A generally accepted solution to this is to use error-correcting logic qubits instead. These include distributing individual qubits across a collection of hardware qubits so that an error in any one of these qubits does not completely destroy the information.
Adding qubits allows you to add error correction to these logical qubits. These are linked to the hardware qubits that hold the logical qubits, allowing their state to be monitored in a way that identifies when an error occurs. These additional qubits can be manipulated to restore the state that was lost when an error occurs.
In theory, this error correction allows the hardware to retain quantum states much longer than individual hardware qubits can.
The trade-off is a significant increase in complexity and number of qubits. The latter is obvious. If each logical qubit requires tens of qubits, many more hardware qubits are needed to run the algorithm. Complete error correction also requires repeated measurements to determine when the error occurred, identify the type of error, and perform the necessary corrections. And all of this must be done while the logical qubits are also being used to run those algorithms.
There’s also the practicality of making this work. Understanding how to perform operations on a pair of hardware qubits is very easy (by a very loose definition of “easy”). It’s much harder to figure out how to do that if each individual hardware qubit only holds at most a fraction of a logical qubit. Adding to the complexity is that a variety of potential error correction schemes exist, and their trade-offs in terms of robustness, convenience, and qubit usage are still being worked out.
That doesn’t mean there hasn’t been progress. Error-corrected qubits have been demonstrated that preserve quantum information better than the hardware qubits that host them. Also, in some cases, individual quantum operations (called gates) have been demonstrated using pairs of logical qubits. And two companies (Atom Computing and IBM) have been increasing the number of qubits to provide enough hardware to host large numbers of logical qubits.
Enter QuEra
Similar to Atom Computing, QuEra’s hardware uses neutral atoms, which has several advantages. Quantum information is stored in the nuclear spins of individual atoms, which are relatively stable in terms of maintaining quantum information. Also, because all atoms of a particular isotope are equivalent, there is no variation between devices, as there is in qubits based on superconducting hardware. Individual atoms can be addressed with lasers without the need for wiring, and atoms can be moved, potentially linking any qubit to other qubits.
QuEra’s current generation of hardware supports up to 280 atom-based qubits. To make this work, we moved these atoms between several functional regions. One is simple storage, which exists when the qubit is not being manipulated or measured. It maintains both the logical qubits in use and a pool of unused qubits that are available during the execution of the algorithm. There is also an “entanglement zone” where these operations take place, and a readout zone where the state of individual qubits can be measured without disturbing qubits elsewhere in the hardware.
