This article is more than **1 year old**

# IBM forges entanglement to double quantum simulations by 'cutting up a larger circuit into smaller circuits'

## Littler circuits executing on smaller hardware and tolerating 'a lot more noise'

IBM says it has found a way to solve problems using fewer qubits than before, effectively doubling the capability of a quantum system by combining both quantum and classical resources.

These claims come in a recently published research paper, in which an IBM team demonstrated what it calls "entanglement forging" to simulate the ground state energy of a water molecule, representing 10 spin-orbitals using just five qubits of a quantum processor rather than 10. A spin-orbital is a wave function that covers both the position and spin angular momentum of a single particle.

Entanglement forging, it turns out, involves the use of a classical computer to capture quantum correlations and effectively split the problem in half, making it possible to separate the 10 spin-orbitals of the into two groups of five that could be processed separately. This doubles the size of the system that can be simulated on quantum hardware.

In the paper, published in open-source journal PRX Quantum, the IBM research team describes how they were able to successfully represent the ground energy state of a water molecule using just five qubits of IBM's 27-qubit Falcon quantum processor. Entanglement forging could markedly expand the computational power of quantum systems, IBM claimed.

This is likely to be a small step on the road to making quantum computers that are capable of solving complex real-world problems. It has been estimated that quantum hardware will need thousands of qubits to solve many practical problems, and will also need to become much more reliable.

However, entanglement forging is a particularly scalable method, according to Sarah Sheldon, one of the IBM Quantum researchers who co-authored the paper. That holds for problems involving weak entanglement, at least.

Entanglement forging can also be applied to systems that are not weakly entangled, but that means doing more legwork on the classical computer to determine how best to partition the system, or to represent the correlation between the two halves, she explained.

The entanglement forging technique essentially involves dividing the system being simulated into two weakly entangled halves, modelling those halves separately on a quantum computer, and then using classical resources to calculate the entanglement between them.

In this case, the two halves correspond to the spin-up and spin-down parts of the molecule.

The outcomes of the two separate halves are fed into a summation that is weighted by a list of values that determine the entanglement structure of the original system, representing the correlations between the two halves.

A classical computer is used in representing the entanglement structure between the two halves by keeping track of this list of values, and those values then determine the smaller experiments the quantum computer must run to calculate the properties of the entire state, IBM said.

Naturally, IBM believes that this technique of entanglement forging will prove useful in solving quantum problems in the near future.

- Fujitsu wants technology to shape a better future – its technology, of course
- 2021 in storage: We waited for a flash price revolution that never came. But about creativity? We can't complain
- Could we use an LLVM-based cross-compiler to build apps for quantum computers? This alliance says yes
- Quantum computing to grow by 50 per cent per year until 2027, when revenue will still be chump change

"Entanglement forging essentially enables you to cut up a larger circuit into smaller circuits that we can execute on smaller hardware," IBM Quantum platform lead Blake Johnson said in a statement.

"Smaller circuits aren't just easier to execute. They're also able to tolerate a lot more noise just by virtue of being smaller."

Meanwhile, progress continues on enlarging quantum systems. IBM's 27-qubit Falcon processor dates from 2019, and has since been surpassed by larger systems, including IBM's own 127-qubit Eagle last year. As detailed at the time, IBM intends to use that design to scale to a 433-qubit processor called Osprey this year, and a 1,121-qubit processed called Condor in 2023.

Investors also continue to pour money into firms developing quantum systems. A firm called Atom Computing has recently scored a $60M Series B funding round after announcing its 100-qubit quantum computer, for example, while Swiss startup Terra Quantum also raised $60m in a Series A funding round. ®

6