Session J3: Invited Session: Quantum Computing with Superconducting Circuits

Continuous High-Fidelity Monitoring of a Superconducting Qubit: From Quantum Jumps to Feedback

Great advances have been made in superconducting qubit technology since the first demonstration of coherent oscillations more than 10 years ago. Continuous, high-fidelity monitoring of the qubit state, however, has remained an elusive target. We demonstrate this functionality by using a wide-bandwidth superconducting parametric amplifier operating near the quantum noise limit to read out the state of a transmon qubit coupled to a linear resonant cavity. Depending on the qubit-cavity detuning and the cavity photon occupation, two different measurement regimes can be accessed. In the strong measurement regime, the qubit states are fully resolved in a time much shorter than $T_{1}$. This permits the observation of quantum jumps between qubit states in real time, and enables the study of the quantum Zeno effect and measurement non-idealities. In the weak measurement regime, information is extracted slowly and the measurement is no longer projective on short time scales. However, the measurement record is still highly correlated with the qubit dynamics and can be used to steer the qubit state using feedback. We demonstrate this idea by phase-locking Rabi oscillations to a classical reference. This allows the oscillations to persist indefinitely, albeit with a reduced amplitude indicative of the efficiency of the feedback protocol.   

Observation of the dynamical Casimir effect in a superconducting circuit

Modern quantum theory predicts that the vacuum of space is not empty, but instead teeming with virtual particles flitting in and out of existence. While initially a curiosity, it was quickly realized that these vacuum fluctuations had measurable consequences, for instance producing the Lamb shift of atomic spectra and modifying the magnetic moment for the electron. This type of renormalization due to vacuum fluctuations is now central to our understanding of nature. 40 years ago, Moore suggested that a mirror undergoing relativistic motion could convert virtual photons into directly observable real photons. This effect was later named the dynamical Casimir effect (DCE). Using a superconducting circuit, we have observed the DCE for the first time. The circuit consists of a coplanar transmission line with an \textit{electrical length} that can be changed at a substantial fraction of the speed of light. The length is changed by modulating the inductance of a superconducting quantum interference device (SQUID) at high frequencies ($> 10$ GHz). In addition to observing the creation of real photons, we observe two-mode squeezing of the emitted radiation, which is a signature of the quantum character of the generation process.   

Towards a scalable superconducting qubit architecture

I will review IBM's current approach towards building a scalable superconducting qubit architecture. The goal is to build a system using quantum error correction schemes based on two-dimensional surface codes, which are predicted to have a remarkably low fault tolerant threshold. I will briefly outline the particular skew-square lattice configuration and describe the concept of logical qubits and gates. On the experimental side, I will show recent advances towards implementing such surface code. This includes improvements in qubit coherence times, now exceeding T1=5us reliably, achieved by shielding the sample from infrared radiation confirming other recent results. Additional approaches leading to even longer coherence times are reviewed. With improved coherence times we show improved gate fidelities of two-qubit gates based the cross-resonance effect -- an all microwave approach towards implementing two-qubit gates. I will conclude by reviewing some interesting future engineering challenges that should be addressed on our way towards building a quantum computer.   

The Photon Shell Game and the Quantum von Neumann Architecture with Superconducting Circuits

Superconducting quantum circuits have made significant advances over the past decade, allowing more complex and integrated circuits that perform with good fidelity. We have recently implemented a machine comprising seven quantum channels, with three superconducting resonators, two phase qubits, and two zeroing registers. I will explain the design and operation of this machine, first showing how a single microwave photon $| 1 \rangle$ can be prepared in one resonator and coherently transferred between the three resonators. I will also show how more exotic states such as double photon states $| 2 \rangle$ and superposition states $| 0 \rangle + | 1 \rangle$ can be shuffled among the resonators as well [1]. I will then demonstrate how this machine can be used as the quantum-mechanical analog of the von Neumann computer architecture, which for a classical computer comprises a central processing unit and a memory holding both instructions and data. The quantum version comprises a quantum central processing unit (quCPU) that exchanges data with a quantum random-access memory (quRAM) integrated on one chip, with instructions stored on a classical computer. I will also present a proof-of-concept demonstration of a code that involves all seven quantum elements: (1), Preparing an entangled state in the quCPU, (2), writing it to the quRAM, (3), preparing a second state in the quCPU, (4), zeroing it, and, (5), reading out the first state stored in the quRAM [2]. Finally, I will demonstrate that the quantum von Neumann machine provides one unit cell of a two-dimensional qubit-resonator array that can be used for surface code quantum computing. This will allow the realization of a scalable, fault-tolerant quantum processor with the most forgiving error rates to date. \\[4pt] [1] M. Mariantoni \textit{et al.}, Nature Physics \textbf{7}, 287-293 (2011.)\\[0pt] [2] M. Mariantoni \textit{et al.}, Science \textbf{334}, 61-65 (2011).   

Realization of Three-Qubit Quantum Error Correction with Superconducting Circuits

Quantum computers promise to solve certain problems exponentially faster than possible classically but are challenging to build because of their increased susceptibility to errors. Remarkably, however, it is possible to detect and correct errors without destroying coherence by using quantum error correcting codes. The simplest of these are the three-qubit codes, which map a one-qubit state to an entangled three-qubit state and can correct any single phase-flip or bit-flip error of one of the three qubits, depending on the code used [1]. The fidelity of a process in which errors can occur on all qubits, where there is the possibility of an uncorrectable double or triple error, should therefore decrease only quadratically with error probability. I will first introduce how the three-qubit encoded state can be produced in our superconducting architecture by employing interactions with non-computational qubit states, as previously demonstrated [2]. I will then discuss how these non-computational interactions can be generalized to produce a novel three-qubit conditional-conditional NOT (CCNot) or Toffoli gate, which implements the correcting step of an error correction algorithm. Finally, I will explain how, by combining these ingredients, we have performed a single pass of both quantum bit- and phase-flip error correction and have demonstrated the predicted first-order insensitivity to errors.\\[4pt] [1] M. A. Nielsen and I. L. Chuang, Cambridge University Press, 2000.\\[0pt] [2] L. DiCarlo, et al. Nature 467, 574 (2010).