Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
Qubits or quantum bits are the fundamental building block for quantum information processes. Whereas conventional computers store and process data as a series of ‘1’s and ‘0’s, quantum computers use the properties of a quantum system, such as the polarization of a photon or the spin of an electron.
Ensuring high-fidelity quantum gates while increasing the number of qubits poses a great challenge. Here the authors present a scalable strategy for optimizing frequency trajectories as a form of error mitigation on a 68-qubit superconducting quantum processor, demonstrating high single- and two-qubit gate fidelities.
Understanding machine learning models’ ability to extrapolate from training data to unseen data - known as generalisation - has recently undergone a paradigm shift, while a similar understanding for their quantum counterparts is still missing. Here, the authors show that uniform generalization bounds pessimistically estimate the performance of quantum machine learning models.
A micro-fabricated Penning trap that operates at a 3 T magnetic field demonstrates full quantum control of an ion and the ability to transport the ion arbitrarily in the trapping plane above the chip.
Ground state representations with artificial neural network methods enable high-accuracy simulations of quantum many-body systems. The authors study the performance of the transformer network architecture on this task and demonstrate its vast potential for novel findings in quantum physics.
Despite being essential to many applications in quantum science, entanglement can be easily disrupted by decoherence. A protocol based on repetitive quantum error correction now demonstrates enhanced coherence times of entangled logical qubits.
Despite recent breakthroughs in quantum error correction experiments with trapped ions, superconducting circuits and reconfigurable atom arrays, there are still several technological challenges to overcome.
A robust initialization protocol has been demonstrated for a four-qubit nuclear spin register in silicon. The protocol, driven electrically through electric-dipole spin resonance, enables high-fidelity qubit control and hence a route to a register-based quantum computer that exploits the exceptional coherence properties of atom qubits in silicon.
Electrons trapped above the surface of solid neon can be used to create qubits using spatial states with different charge distributions. These charge qubits combine direct electric field control with long coherence times.
Measurements of two neighbouring silicon-based qubits show that the charge noise they each experience is correlated, suggesting a common origin. Understanding these correlations is crucial for performing error correction in these systems.
A real qubit is not an isolated unitary quantum system but is subject to noise from its environment. An experiment has now turned this interaction on its head, controlling the environment using the qubit itself.