Extensive characterization of a family of efficient three-qubit gates at the coherence limit

TL;DR

This paper introduces a family of efficient three-qubit gates implemented on superconducting qubits, achieving high fidelity near the coherence limit and enabling faster, more flexible quantum circuit construction.

Contribution

The authors demonstrate a new three-qubit gate method that extends to various hardware, improves speed, and enhances circuit flexibility, with near-coherence-limited fidelity.

Findings

Achieved 97.90% gate fidelity near the device coherence limit.

Generated GHZ and W states with a single application of the new gate.

Compared to standard decompositions, the new gate reduces circuit depth for entangled states.

Abstract

While all quantum algorithms can be expressed in terms of single-qubit and two-qubit gates, more expressive gate sets can help reduce the algorithmic depth. This is important in the presence of gate errors, especially those due to decoherence. Using superconducting qubits, we have implemented a three-qubit gate by simultaneously applying two-qubit operations, thereby realizing a three-body interaction. This method straightforwardly extends to other quantum hardware architectures, requires only a "firmware" upgrade to implement, and is faster than its constituent two-qubit gates. The three-qubit gate represents an entire family of operations, creating flexibility in quantum-circuit compilation. We demonstrate a gate fidelity of $97.90\%$, which is near the coherence limit of our device. We then generate two classes of entangled states, the GHZ and W states, by applying the new gate only once; in comparison, decompositions into the standard gate set would have a two-qubit gate depth of two and three, respectively. Finally, we combine characterization methods and analyze the experimental and statistical errors on the fidelity of the gates and of the target states.