Techniques for error correction of a logical qubit and related systems and methods

Abstract

Techniques for quantum error correction of a multi-level system are provided and described. In some aspects, techniques for encoding a state of a multi-level quantum system include encoding a quantum information in a two-mode state of two quantum mechanical oscillators. Techniques for protecting the two-mode state against dephasing and energy loss are described.

Description

BACKGROUND

Quantum information processing techniques perform computation by manipulating one or more quantum objects. These techniques are sometimes referred to as “quantum computing.” In order to perform computations, a quantum information processor utilizes quantum objects to reliably store and retrieve information. According to some quantum information processing approaches, a quantum analogue to the classical computing “bit” (being equal to 1 or 0) has been developed, which is referred to as a quantum bit, or “qubit.” A qubit can be composed of any quantum system that has two distinct states (which may be thought of as 1 and 0 states), but also has the special property that the system can be placed into quantum superpositions and thereby exist in both of those states at once.

Several different types of qubits have been successfully demonstrated in the laboratory. However, the lifetime of the states of some types of systems before information is lost due to decoherence of the quantum state, or to other quantum noise, may be on the order of ˜100 μs. Notwithstanding longer lifetimes, it may be important to provide error correction techniques in quantum computing that enable reliable storage and retrieval of information stored in a quantum system. However, unlike a classical computing system in which bits can be copied for purposes of error correction, the no-cloning theorem precludes the cloning of an unknown state of a quantum system. The system may, however, be entangled with other quantum systems which effectively spreads the information in the system out over several entangled objects.

SUMMARY

According to some aspects, a system is provided comprising a circuit quantum electrodynamics system that includes a logical qubit comprising first and second quantum mechanical oscillators, a mixing circuit coupled to each of the first and second quantum mechanical oscillators, and a quantum error syndrome detection circuit coupled to each of the first and second quantum mechanical oscillators, at least one controller, and at least one computer readable medium storing instructions that, when executed by the at least one controller, perform a method comprising applying, using the at least one controller, a plurality of drive waveforms to the mixing circuit, the plurality of drive waveforms having different frequencies, and concurrent with said application of a plurality of drive waveforms, detecting, by the quantum error syndrome detection circuit, a change in energy state of the logical qubit.

According to some embodiments, the plurality of drive waveforms are configured to drive a dissipation of energy from the first and second quantum mechanical oscillators to the mixing circuit.

According to some embodiments, the mixing circuit is coupled to the first and second quantum mechanical oscillators via a first Josephson junction.

According to some embodiments, the plurality of drive waveforms are configured to produce an exchange of energy between the first and second quantum mechanical oscillators and the first Josephson junction.

According to some embodiments, said exchange of energy is configured to passively protect against dephasing errors of the logical qubit.

According to some embodiments, the first and second quantum mechanical oscillators exhibit different resonant frequencies.

According to some embodiments, the first and second quantum mechanical oscillators each comprise a microwave cavity.

According to some embodiments, the quantum error syndrome detection circuit comprises a fourth quantum mechanical oscillator independently coupled to each of the first and second quantum mechanical oscillators.

According to some embodiments, the quantum error syndrome detection circuit further comprises second and third Josephson junctions, the fourth quantum mechanical oscillator is coupled to the first quantum mechanical oscillator via the second Josephson junction, and the fourth quantum mechanical oscillator is coupled to the second quantum mechanical oscillator via the third Josephson junction.

According to some embodiments, the instructions are further configured to, when executed by the at least one controller, apply first and second pumps to the second and third Josephson junctions, respectively.

According to some embodiments, the instructions are further configured to, when executed by the at least one controller, detect an energy displacement of the fourth quantum mechanical oscillator caused by application of the first and second pumps, wherein said energy displacement is indicative of the change in the energy state of the logical qubit.

According to some embodiments, the quantum error syndrome detection circuit comprises a fourth Josephson junction coupled to each of the first and second quantum mechanical oscillators.

According to some embodiments, the change in the energy state of the logical qubit is implicitly detected via application of one or more drive waveforms to the first quantum mechanical oscillator that produce a change in the energy state of the first quantum mechanical oscillator which is dependent upon a state of the fourth Josephson junction.

According to some aspects, a method of performing quantum error correction is provided within a circuit quantum electrodynamics system that includes a logical qubit comprising first and second quantum mechanical oscillators, a mixing circuit coupled to each of the first and second quantum mechanical oscillators, and quantum error syndrome detection circuit coupled to each of the first and second quantum mechanical oscillators, the method comprising applying, using at least one controller, a plurality of drive waveforms to the mixing circuit, the plurality of drive waveforms having different frequencies, and concurrent with said application of a plurality of drive waveforms, detecting, by the quantum error syndrome detection circuit, a change in energy state of the logical qubit.

According to some embodiments, the plurality of drive waveforms drive a dissipation of energy from the first and second quantum mechanical oscillators to the mixing circuit.

According to some embodiments, the plurality of drive waveforms produce an exchange of energy between the first and second quantum mechanical oscillators and a first Josephson junction, and wherein the mixing circuit is coupled to the first and second quantum mechanical oscillators via the first Josephson junction.

According to some aspects, a circuit quantum electrodynamics system is provided comprising a physical qubit, a logical qubit comprising a first bosonic system and a second bosonic system, the first and second bosonic systems being coupled to the physical qubit, wherein a state of the physical qubit is encoded in a state of the logical qubit that is a multi-component coherent state of the bosonic system, and an error syndrome detector coupled to the logical qubit and configured to detect an error in said state of the logical qubit based on one or more quantum non-destructive measurements of a boson number difference between the first and second bosonic systems.

According to some embodiments, the error syndrome detector comprises a third bosonic system independently coupled to each of the first and second bosonic systems.

According to some embodiments, the error syndrome detector further comprises second and third Josephson junctions, wherein the third bosonic system is coupled to the first bosonic system via the second Josephson junction, and wherein the third bosonic system is coupled to the second bosonic system via the third Josephson junction.

According to some embodiments, the error syndrome detector is further configured to detect an energy displacement of the third bosonic system caused by application of energy to the third bosonic system, wherein said energy displacement is indicative of the boson number difference between the first and second bosonic systems.

The foregoing apparatus and method embodiments may be implemented with any suitable combination of aspects, features, and acts described above or in further detail below. These and other aspects, embodiments, and features of the present teachings can be more fully understood from the following description in conjunction with the accompanying drawings.

BRIEF DESCRIPTION OF DRAWINGS

Various aspects and embodiments will be described with reference to the following figures. It should be appreciated that the figures are not necessarily drawn to scale. In the drawings, each identical or nearly identical component that is illustrated in various figures is represented by a like numeral. For purposes of clarity, not every component may be labeled in every drawing.

FIGS. 1A-1B illustrate encoding of the state of a multi-level quantum system in a bosonic system, according to some embodiments;

FIG. 2 depicts an illustrative system for encoding a logical qubit in the state of two quantum mechanical oscillators, according to some embodiments;

FIG. 3 depicts an illustration of the Bloch sphere of a qubit encoded with the pair-cat codes, according to some embodiments;

FIG. 4 depicts a plurality of subspaces of the pair-cat codes, according to some embodiments;

FIG. 5 depicts an illustrative system for encoding a logical qubit in the state of two quantum mechanical oscillators, according to some embodiments;

FIG. 6 illustrates a three-dimensional energy level diagram illustrating a dissipative process which protects against dephasing errors, according to some embodiments;

FIG. 7 illustrates frequencies of drive waveforms which may be applied to produce the dissipative process of FIG. 6, according to some embodiments;

FIG. 8 illustrates a principle of applying photon number-dependent displacements to detect a photon loss error, according to some embodiments;

FIGS. 9A-9C depict an illustrative system for encoding a logical qubit in the state of two quantum mechanical oscillators, according to some embodiments; and

FIG. 10 is a block diagram of a circuit quantum electrodynamics system, according to some embodiments.

DETAILED DESCRIPTION

The present application relates to an improved technique for correcting errors in the state of a multi-mode quantum system. An “error” in the quantum error correction context refers to a change in the state of the quantum system that may be caused by, for instance, energy (e.g., photon) losses, energy gains, dephasing, time evolution of the system, etc., and which alters the state of the system such that the information stored in the system is altered.

As discussed above, quantum multi-level systems such as qubits exhibit quantum states that may decohere too quickly for practical quantum computation. While experimental techniques will undoubtedly improve on this and produce qubits with longer decoherence times, it may nonetheless be beneficial to couple a multi-level system to another system that exhibits much longer decoherence times. When the information stored by a qubit is stored within a coupled system, the coupled system is often referred to as a “logical qubit” because it stores information like a qubit though it does so indirectly and not through a single multi-mode state (by way of comparison, the solitary single multi-mode state qubit may be termed a “physical qubit”).

As will be described further below, it may be particularly desirable to produce a logical qubit from one or more bosonic modes (e.g., modes of one or more microwave cavities). Generally speaking, a bosonic mode may be electromagnetic, magnetic, or even mechanical oscillator modes, although, in the current context, it is photonic and relates to number states of photons stored in a cavity. While quantum information stored in bosonic modes may have a longer lifetime than a qubit state, the lifetime may still be limited and errors may still occur within the bosonic system. It may therefore be desirable to arrange a bosonic system so that when errors in its state occur, these errors can be effectively corrected and the prior state of the system may thereby be regained. If a broad class of errors can be corrected for, it may be possible to maintain the state of the bosonic system indefinitely (or at least for long periods of time) by correcting for any type of error that might occur.

The fields of cavity quantum electrodynamics (cavity QED) and circuit QED represent one illustrative experimental approach to implement quantum error correction. In these approaches, quantum information may be encoded in the state of a qubit, which may be a physical qubit or a logical qubit.

In some approaches, one or more physical qubits are each coupled to a logical qubit in such a way as to allow mapping of the quantum information contained in the physical qubit to and/or from the logical qubit. The logical qubits generally will have longer stable lifetimes than the physical qubits. The quantum state may later be retrieved in a physical qubit by mapping the state back from a respective logical qubit to the physical qubit.

When a multi-level system, such as a physical qubit, is mapped onto the state of a bosonic system to which it is coupled, a particular way to encode the qubit state in the bosonic system must be selected. This choice of encoding is often referred to simply as a “code.”

As an example, a code might represent the ground state of a qubit using the zero boson number state of a resonator and represent the excited state of a qubit using the one boson number state of the resonator. That is:(c_g|g+c_e|e)⊗|0→|g⊗(c_g|0+c_h|1)where |g is the ground state of the qubit, |e is the excited state of the qubit, c_g and c_e are complex numbers representing the probability amplitude of the qubit being in state |g or |e, respectively, and |0 and |1 are the zero boson number state and one boson number state of the resonator, respectively. While this is a perfectly valid code, it fails to be robust against many errors, such as boson loss. That is, when a boson loss occurs, the state of the resonator prior to the boson loss may be unrecoverable with this code.

The use of a code can be written more generally as:(c_g|g+c_e|e)⊗|0→|g⊗(c_g|W_↓+c_e|W_↑)where ⊗ signifies a tensor product, the right arrow signifies a code mapping, and |W_↓and |W_↑are referred to as the logical codewords (or simply “codewords”). The choice of a code—equivalently, the choice of how to encode the state of a two-level system (e.g., a qubit), representing one bit of quantum information, in the state of the bosonic system—may therefore comprise choosing values for |W_↓and |W₅₂. FIGS. 1A-1B graphically depict this process of encoding for some choice of |W_↓and |W_↑.

When an error occurs, the state of the logical qubit transforms to a superposition of resulting states, herein termed “error words,” |E_↓^k and |E_↑^k as follows:α|W_↓+β|W_↑→α|E_↓^k+β|E_↑^kwhere the index k refers to a particular error that has occurred. As discussed above, examples of errors include boson loss, boson gain, dephasing, amplitude damping, etc.

In the above example, the logical qubit is used to encode the state of a physical qubit. Alternately, the logical qubit may be used independently to encode a state using a particular code to represent a multi-level system.

Irrespective of whether the state of the logical qubit is encoded based on a coupled physical qubit or as an independent state, an error in the logical qubit may be detected by making one or more measurements of the bosonic system and/or of one or more systems to which the bosonic system is coupled. A particular result of measurements that indicates an error has occurred is often referred to as an “error syndrome.” For instance, certain measurements may indicate that the bosonic system has lost a boson. The presence of an error, and in some cases the type of error, indicated by the error syndrome may imply that the system has changed to some new state in such a way that the error words are known. In some cases, the error words may be determined by implication based on the error syndrome. For instance, as discussed below, with conventional cat codes, parity is measured as an error syndrome of boson loss, which allows an operation to be identified that will return the system to its state prior to the loss of the boson.

In general, the choice of code affects how robust the system is to errors. That is, the code used determines to what extent a prior state can be faithfully recovered when an error occurs. A desirable code would be associated with a broad class of errors for which no information is lost when any of the errors occurs and any quantum superposition of the logical codewords can be faithfully recovered. Some codes, while robust against certain errors, may however be impractical to realize in a physical system.

One class of codes, referred to as “cat codes,” have a desirable property that the state of a bosonic system storing a bit of quantum information using these codes is robust against boson loss errors. Cat codes are described in Leghtas et al., “Hardware-Efficient Autonomous Quantum Memory Protection,” Phys. Rev. Lett., 111(12):120501, 2013, incorporated herein by reference. In particular, the cat codes represent a set of logical codewords that encode a bit of quantum information in such a way that the loss of a boson produces error words that are also part of the same set of codewords. Furthermore, by non-destructively measuring a change in parity of the bosonic system, boson losses can be identified and corrected by transforming the state of the bosonic system back to the state prior to the loss.

In the description below, for purposes of illustration, photon systems such as optical or microwave cavities are discussed, although it will be appreciated that any of the below description that relates to photon systems may equally apply to other suitable bosonic systems and is not limited to the particular systems described.

Using the cat codes, one bit of quantum information (e.g., the state of a physical qubit) may be represented by a multi-component coherent state of a bosonic oscillator:|ψ_α⁽⁰⁾=c_g|C_α⁺+c_e|C_iα⁺where|C_α^±=(|α±|−α),|C_iα^±=(|iα±|−α), is a normalizing factor and |α denotes a coherent state of complex amplitude α, chosen such that |α, |−α, |iα, |−iα are quasi-orthogonal. Together with |ψ_α⁽⁰⁾, the following states may also be introduced:|ψ_α⁽¹⁾=c_g|C_α⁻+ic_e|C_iα⁻|ψ_α⁽²⁾=c_g|C_α⁺−c_e|C_iα⁺|ψ_α⁽³⁾=c_g|C_α⁻−ic_e|C_iα⁻

These states have useful properties. First, when a boson is lost from the system (a type of error as discussed above), the states |ψ_α⁽ⁿ⁾ evolve to another one of the states |ψ_α⁽ⁿ⁾. That is, the set {|ψ_α⁽ⁿ⁾} is closed under the action of the annihilation operator a. Second, in the absence of boson losses, the state deterministically evolves in a predictable manner. As a result, a unitary transformation can be found that is independent of c_g and C_e and which maps an evolved state that may or may not be the result of a boson loss back to the original state, thereby undoing the effect of decoherence. A boson loss can be identified, for instance, through a quantum non-destructive (QND) measurement of boson number parity.

The inventors have recognized and appreciated that, while cat codes may be robust against the loss of a single boson, these codes are not robust against the simultaneous loss of two bosons. That is, the operator a² acting on |ψ_α⁽ⁿ⁾ may produce a state from which the prior state cannot be recovered.

The inventors have further recognized and appreciated techniques for realizing a class of codes, referred to herein as pair-cat codes, that are robust against multiple boson losses. The pair-cat codes described herein encode a bit of quantum information using the combined state of two bosonic systems, and the inventors have realized techniques for correcting errors that may occur as a result of changes in either or both systems. In particular, the techniques comprise a driven-dissipative process that stabilizes the systems' states with respect to dephasing errors, and comprise a process for detecting and correcting any number of non-simultaneous boson-loss errors from either or both systems.

FIG. 2 depicts an illustrative system for encoding a logical qubit in the state of two quantum mechanical oscillators such that the quantum information encoded by the state may be protected against errors, according to some embodiments. In the example of system 200, oscillators 211 and 212 may be arranged with a two-mode quantum state according to the pair-cat codes, which will be described below. The two-mode state of the oscillators may encode a logical qubit 210 such that the quantum information is robust against dephasing and boson-loss errors. In particular, the mixing circuit 220, through application of drive waveforms 225, allows dissipation of energy from the oscillators 211 and 212 via the coupling 215 in such a way as to protect against dephasing errors in the two-mode state. For instance, the mixing circuit 220 may comprise one or more elements configured to convert the drive waveforms 225 into operations upon the logical qubit 210. In the example of FIG. 2, the error syndrome detection circuit 240 is configured to detect boson losses from either or both oscillators 211 and 212 (i.e., from either or both modes of the two-mode state). In some embodiments, the error syndrome detection circuit 240 may also be configured to perform actions to change a state of the two-mode state of oscillator 211 and oscillator 212 and thereby correct for detected boson-loss errors.

According to some embodiments, system 200 may protect the quantum information encoded by logical qubit 210 through passive protection. Passive protection, as referred to herein, comprises an ongoing process that corrects for one or more errors without it being necessary to measure an error syndrome and determine an appropriate recovery operation based on the measurement. Such processes may also be referred to as continuous quantum error correction (QEC).

In order to describe the manner in which the system of FIG. 2 protects the logical qubit 210 against various errors, the pair-cat codes are explained below. For convenience, photon systems are considered in the below discussion, although, as discussed above, the below discussion may equally be applied to any other bosonic system. Let [F] be a dissipator:[F](ρ)=FρF^†−½{F^†F,ρ}where F is a jump operator and p a density matrix. In the conventional cat code approach described above, a stabilizing jump operator F_I=a⁴−a⁴ is considered, which annihilates all steady states and which commutes with photon number parity.

In the case of the pair-cat codes, the stabilizing jump operator F_II=a² b²−γ⁴ is considered, where in the pair-cat codes the annihilation operators of the two oscillators are a and b and complex parameters γ are used to denote the pair-cat states. Each oscillator mode obeys the standard commutation relations [a, a^†]=[b, b^†]=1 and [a, b^†]=0 and photon number operators for each oscillator are denoted {circumflex over (n)}=a^†a and {circumflex over (m)}=b^†b.

Time evolution of the two-mode density matrix ρ is governed by the Lindbladian{dot over (ρ)}=κ₂₂(ρ)+ . . .where ₂=[F_II]. κ₂ is a non-negative rate and “ . . . ” represents competing error processes, which include loss errors and dephasing errors. The loss errors are caused by dissipators of the form κ_a[a] and κ_b[b], and the dephasing errors caused by dissipators of the form κ_n[{circumflex over (n)}] and κ_m[{circumflex over (m)}]. Quantum information may be encoded in steady states of ₂, which are states ρ in which ₂ (ρ)=0. These states form a decoherence-free subspace of ₂ represented by its projection P₂.

A basis may be constructed for the two-mode coherent states using the difference in photon number between the two modes Δ, as determined by the operator {circumflex over (Δ)}={circumflex over (m)}−{circumflex over (n)}. The operator {circumflex over (Δ)} commutes with ab and therefore commutes with F_II. Thus, the space of states annihilated by the jump can be spanned by a basis of states with fixed eigenvalues Δ∈. To determine these states, let us define projections onto sectors of fixed Δ,

$P_{\Delta }=\{\begin{array}{cc}\sum _{n=0}^{\infty }n,n+\Delta \rangle \langle n,n+\Delta & \Delta \ge 0\\ \mathrm{SWAP}{P}_{\left|\Delta \right|}\mathrm{SWAP}& \Delta <0\end{array}$ where the SWAP operator SWAP |n, m=|m, n is

$\mathrm{SWAP}=\exp \left[i\frac{\pi }{2}\left(a^{†}-b^{†}\right)\left(a-b\right)\right].$

From here on for convenience it is assumed that Δ≥0, since an application of SWAP will produce corresponding results for Δ<0. The above projections P_Δ may be applied to the two-mode coherent state |γ, γ to determine a basis for the code space. Such projection yields:

$\begin{array}{c}{\gamma }_{\Delta }\rangle =\frac{P_{\Delta }\gamma ,\gamma \rangle }{\sqrt{N_{\Delta }}}\\ =\frac{1}{\sqrt{I_{\Delta }\left(2{\gamma }^2\right)}}\sum _{n=0}^{\infty }\frac{{\gamma }^{2n+\Delta }}{\sqrt{n!\left(n+\Delta \right)!}}n,n+\Delta \rangle \end{array}$ with I_Δ being a modified Bessel function of the first kind and normalization:N_Δ=γ,γ|P_Δ|γ,γ=e^−2|γ|2I_Δ(2|γ|²)

It can be shown, since ab commutes with P_Δ, that ab|γ_Δ=γ²|γ_Δ).

An illustration of how a qubit may be encoded with the pair-cat codes is shown in FIG. 3. Note that, since |γ_Δ is dependent on Δ, the encoding is specific to a particular Δ subspace. Bloch sphere 300 depicts a geometrical representation of the state space of the two-mode logical qubit represented by oscillators 211 and 212 shown in FIG. 2.

With respect to errors occurring when quantum information is encoded in logical qubit 210, initially it is noted that a and b by shift the number difference Δ in opposite directions:aP_Δ=P_Δ+1a bP_Δ=P_Δ−1b

That is, a photon jump changes the A subspace but retains the projection within the subspace. For instance, given logical basis states within each subspace of fixed Δ,|0_Δp=N_Δ(|γ_Δ+i^Δ|iγ_Δ)|1_Δp=N_Δ(|γ_Δ+i^Δ|iγ_Δ)any number of photon jumps produce:aⁿ|0_Δp=|0_Δ+np b^m|0_Δp=|0_Δ−mp

A photon jump in either oscillator simply changes the Δ subspace and is correctable. For instance, in the system of FIG. 2, error syndrome detection circuit 240 may be configured to detect a change in the number difference (e.g., by performing a QND measurement of Δ). The error syndrome detection circuit 240 and/or one or more other components of system 200 not depicted may perform an operation on one or both of oscillators 211 and 212 to return the state to a desired Δ subspace (e.g., the Δ=0 subspace).

FIG. 4 depicts a selection of the Δ subspaces (Δ=−1, 0 and 1) to illustrate how photon jumps simply change the Δ subspace. Note that the conventional cat codes rely on the parity as a quantum number, which has two distinct eigenvalues (odd or even), whereas the pair-cat codes rely on Δ as a quantum number, which has an infinite number of values, and therefore an infinite number of code spaces exist.

As also shown in FIG. 4, a simultaneous loss of photons from both oscillators leads to a bit flip but does not produce an error syndrome with the pair-cat codes, and therefore represents an uncorrectable error:ab|0_Δp=N_Δ(|γ_Δ−i^Δ|iγ_Δ)=|1_Δp

Through frequent detection and correction of photon jump errors, the rate of such errors may be reduced or avoided, however. Since precisely simultaneous loss of photons from both oscillators may not occur, a sufficiently frequent detection and correction may resolve individual jumps and thereby allow correction of all photon loss errors.

Moreover, the dissipator ₂=[F_II]=[a²b²−γ⁴] is protective of dephasing errors. In particular, since [{circumflex over (n)}] and [{circumflex over (m)}] both commute with {circumflex over (Δ)}, dephasing does not connect code spaces for different values of Δ. This leaves dephasing within each Δ subspace, and the rates of such errors decrease exponentially with γ. As a result, dephasing errors may be suppressed by implementing a suitably large value of γ (in the system of FIG. 2, for instance).

In view of the above, by engineering suitable states of the oscillators 211 and 212 in FIG. 2 according to the pair-cat codes, errors may be suppressed as follows. First, through application of drive waveforms 225 to the mixing circuit 220, a dissipative process of the form [a²b²−γ⁴] may be realized and thereby protect against dephasing errors. Second, by detection of photon jump errors using error syndrome detection circuit 240, a suitable operation (or operations) to correct for said jump errors may be applied to the oscillator 211 and/or oscillator 212 by the circuit 240 or otherwise. These two mechanisms may thereby allow for the two-mode state encoded by the oscillators in FIG. 2 to be robust against all types of errors except for a simultaneous loss (or gain) of photons in both cavities.

System 500 shown in FIG. 5 represents one illustrative implementation of the system of FIG. 2, according to some embodiments. In the example of FIG. 5, logical qubit 515 is implemented via cavities 511 and 512, which are each coupled to a Josephson junction 518. The cavity 520 facilitates entropy extraction via the driven-dissipative process as discussed above through a mixing process implemented through drive waveforms 525. Cavity 540 is coupled to cavities 511 and 512 via respective Josephson junctions 541 and 542, and is configured to perform measurement of the error syndrome {circumflex over (Δ)}=b^†b−a^†a, which is the photon number difference in the cavities 511 and 512. Drive waveforms 545 are configured to realize a displacement of cavity 540 that is proportional to {circumflex over (Δ)}.

According to some embodiments, cavities 511 and 512 may be comparatively high-Q cavities, whereas cavities 520 and 540 may be comparatively low-Q cavities. Any one or more of cavities 511, 512, 520 and 540 may, for example, comprise a resonating cavity such as an optical cavity or a microwave cavity.

According to some embodiments, drive waveforms 525 may be configured to introduce an exchange of energy between one or both of cavities 511 and 512 and the Josephson junction 518. In some cases, energy may be transferred from one or both of the cavities 511 and 512 to the junction, thereby exciting the junction. Subsequently, the junction may be reset to a lower energy level, thereby realizing an appropriate dissipative process.

An example of such an approach is shown in FIG. 6, which illustrates a three-dimensional energy level diagram 600 having axes of the number densities of each cavity ({circumflex over (n)}=a^†a and {circumflex over (m)}=b^†b) and of the excitation level of the Josephson junction 518. In FIG. 6, the lowest three eigenstates (energy levels) of the junction 518 are labeled g, e and f in order of increasing energy. The Fock-states of the cavities 511 and 512 are denoted by numbers n and m.

In the example of FIG. 6, a cascading process is applied to an initial state |g, n, m, i.e., with the junction 518 in its ground (g) state and the two cavities 511 and 512 having Fock-states n and m, respectively. In an initial step 601 of the process, the application of energy to the cavity 520 via one or more of the drive waveforms 525 connects the initial state |g, n, m to a virtual state detuned from the state |e, n, m−2 by an amount δ. This detuning is shown in FIG. 6 via the dashed line connecting to the |e, n, m−2 state. In step 602, the detuned |e, n, m−2 state is connected with a state |f, n−2, m−2 through further application of energy to the cavity 520 via one or more of the drive waveforms 525. As a result of the steps 601 and 602, a pair of two-photon exchanges (one from each of the cavities 511 and 512) cause excitation of the junction 518.

In step 603, the effective two-photon dissipation on each of the two cavities may be implemented by having the junction relax from |f, n−2, m−2 to |g, n−2, m−2 through further application of energy to the cavity 520 via one or more of the drive waveforms 525. In step 604 application of energy to the cavity 520 via one or more of the drive waveforms 525 may return the state to |g, n, m. It will be appreciated that any number of drive waveforms 525 may be applied to produce this sequence of state transitions and that any one of the drive waveforms 525 may produce any one or more of the above-described steps.

One non-limiting practical implementation of the process illustrated in FIG. 6 is described below. In this example, a Hamiltonian for the cavities 511 and 512 (with lowering operators a, b and frequencies ω_a and ω_b, respectively) and a mode of the Josephson junction 518 having lowering operator J, frequency ω_J and Josephson energy E_J is considered, driven by a time-dependent drive ℏε(t)J+h.c. LetH′=H₀− 1/24E_Jφ⁴+ℏ(ε(t)J+ε*(t)J^†).comprise the harmonic portion of the full Hamiltonian as well as the anharmonic portion of the junction Hamiltonian, where

$\frac{H_0}{\hslash }={\omega }_a{a}^{†}a+{\omega }_b{b}^{†}b+{\omega }_J{J}^{†}J$ $\hat{\phi }={\varphi }_{a}a+{\varphi }_{b}b+{\varphi }_{J}J+h.c.$ and ϕ_a,b,J=ϕ_ZPF,(a,b,J)/ϕ₀ denote the amplitude participation ratios of the respective modes in the junction with ϕ_ZPF,(a,b,J) corresponding to the zero point fluctuations of the respective modes as seen by the junction and

${\varphi }_0=\frac{\hslash }{2e}$ being the reduced superconducting flux quantum.

Considering driving waveforms that comprise three tones as follows:

$\varepsilon \left(t\right)=\sum _{k=1}^3{ϵ}_{\mathrm{pk}}\exp \left(i{\omega }_{pk}t\right)$ a sequence of transformations may be applied that absorbs, one tone at a time, the entire ε drive into φ⁴, the anharmonic part of the junction 518. The drives ε may be tuned so that the above-described process of FIG. 5 is produced, with the drive frequenciesω_p1=2{tilde over (ω)}_b−{tilde over (ω)}_ge−δω_p2=2{tilde over (ω)}_a−{tilde over (ω)}_ef−δω_p3=½({tilde over (ω)}_ge+{tilde over (ω)}_ef)where the states |e, |f and |g of the junction define a transition frequency {tilde over (ω)}_ge between |g and |e, and a transition frequency {tilde over (ω)}_ef between |e and |f. Moreover, the cavity frequencies ω_a, and ω_b have been (Lamb- and Stark-)shifted to new frequencies {tilde over (ω)}_a, and {tilde over (ω)}_b in producing a new Hamiltonian:

$\frac{{\tilde{H}}_0}{\hslash }={\tilde{\omega }}_a{a}^{†}a+{\tilde{\omega }}_b{b}^{†}b+{\tilde{\omega }}_{\mathrm{ge}}{\hat{\sigma }}_{\mathrm{ee}}+\left({\tilde{\omega }}_{\mathrm{ge}}+{\tilde{\omega }}_{\mathrm{ef}}\right){\hat{\sigma }}_{\mathrm{ff}}$ with σ_kl=|lk| and |l and |k being junction states.

The drive frequencies ω_p1 and ω_p2 are detuned by δ as discussed above to together produce an exchange of two photons in each cavity with the ef excitation of the junction. The drive ω_p3 selects a term {circumflex over (σ)}_gf=|fg| that, in the presence of dissipation, translates into a simultaneous two-photon drive on both cavities and produces the dissipator F_II with nonzero γ. A schematic of the drive frequencies ω_p1 and ω_p2 is shown in FIG. 7.

As discussed above, in the example of FIG. 2 the error syndrome detection circuit 240 and/or one or more other components of system 200 not depicted may perform an operation on one or both of oscillators 211 and 212 to return the state to a desired Δ subspace after the loss of a boson. An illustrative approach to correcting such errors will now be described with respect to the illustrative system 500 of FIG. 5.

In the example of FIG. 5, the error syndrome detection circuit comprises modes of each of the Josephson junctions 541 and 542. These two junction modes couple the cavities 511 and 512 to a cavity 540. The following approach relies upon the four-wave mixing capabilities of the two Josephson junction modes to link the displacement of cavity 540 to the photon number difference between the two modes of the cavities 511 and 512.

At a high level, the approach described below performs displacement operations on cavity 540 that are dependent upon the photon numbers in the cavities 511 and 512. By adjusting the magnitudes and phases of the pumps 545, a total joint cavity dependent displacement may be applied to cavity 540, allowing for direct measurement of the error syndrome {circumflex over (Δ)}. This principle is depicted in the schematic of FIG. 8, which illustrates the simultaneous application of two displacements D(ϵ_Δa^†a) and D(−ϵ_Δb^†b) with pumps 545, which together produce a Δ-dependent displacement of cavity 540.

As shown in FIG. 8, there are two junction modes J_1,2 coupling the two cavities a and b (cavities 511 and 512) to a readout cavity c (cavity 540). It is assumed that cavity a couples only to junction J₁ and cavity b couples only to junction J₂. Both junctions couple to cavity c. It is assumed that the two junction modes are isolated from each other and can be driven independently. The two junctions and cavity c are driven with drives parameterized by ϵ_1,2,c, respectively. Assuming |ℏϵ_1,2,c|<<E_J1,2 and expanding the anharmonic parts of the two junctions yieldsH=H₀− 1/24E_J1{circumflex over (φ)}₁⁴− 1/24E_J2{circumflex over (φ)}₂⁴+ℏexp(i{tilde over (ω)}_ct)(ϵ_cc+ϵ₁J₁+ϵ₂J₂)+h.c. where

$\frac{H_0}{\hslash }={\omega }_a{a}^{†}a+{\omega }_b{b}^{†}b+{\omega }_c{c}^{†}c+{\omega }_{J1}{J}_1^{†}{J}_1+{\omega }_{J2}{J}_2^{†}{J}_2$ is the harmonic part, the phase differences across the junctions k∈{1,2} are{circumflex over (φ)}₁=ϕ_a1a+ϕ_c1c+ϕ₁J₁+h.c., {circumflex over (φ)}₂=ϕ_b2a+ϕ_c2c+ϕ₂J₂+h.c., and ϕ are the amplitude participation ratios.

The two junctions 541 and 542 are both independently driven at the frequency {tilde over (ω)}_c of cavity 540, which we set to be the shifted frequency of mode c after normal ordering. We also apply a direct resonant drive to cavity 540 of strength ϵ_c; the importance of this drive will be clear in the next couple of steps.

We absorb the J_1,2-drives on the junction modes in the respective anharmonicities leave the c-cavity drive untouched. This yieldsH′=H₀− 1/24E_J1[{circumflex over (Φ)}₁(t)]⁴− 1/24E_J2[{circumflex over (Φ)}₂(t)]⁴+ℏϵ_c exp(i{tilde over (ω)}_ct)c+h.c. where the time-dependent phase differences are{circumflex over (Φ)}₁(t)=ϕ_a1a+ϕ_c1c+ϕ₁J₁+ϕ₁ξ₁ exp(i{tilde over (ω)}_ct)+h.c. {circumflex over (Φ)}₂(t)=ϕ_b2b+ϕ_c2c+ϕ₁J₂+ϕ₂ξ₂ exp(i{tilde over (ω)}_ct)+h.c. and ξ_k are the displacements of the junction modes due to the respective drives. Finally, we normal-order the anharmonicities, go into a rotating frame with respect to the Lamb- and Stark-shifted shifted harmonic part,

$\frac{{\tilde{H}}_0}{\hslash }={\tilde{\omega }}_a{a}^{†}a+{\tilde{\omega }}_b{b}^{†}b+{\tilde{\omega }}_c{c}^{†}c+{\tilde{\omega }}_{J1}{J}_1^{†}{J}_1+{\tilde{\omega }}_{J2}{J}_2^{†}{J}_2$ and keep only the non-rotating terms. Since the only drive frequency is {tilde over (ω)}_c, the only off-diagonal time-independent terms are those for which the number of c terms is equal to the number of ξ*_1,2 terms plus the number of c^† terms (and their Hermitian conjugates). We also assume that the junction modes J_1,2 are never resonantly driven and hence are never populated. Therefore, for the sake of compactness, we drop all the diagonal terms involving the J_k^†J_k operator. The system Hamiltonian becomes

$\frac{H_{sys}}{\hslash }=-{\chi }_{ac}{a}^{†}a{c}^{†}c-{\chi }_{bc}{b}^{†}{\mathrm{bc}}^{†}c-\sum _{r=a,b,c}\frac{{\chi }_{\mathrm{rr}}}{2}{r}^{†2}{r}^2-\left({ϵ}_c+g_{dir}+g_{s}c+\sum _{r=a,b,c}{g}_r{r}^{†}r\right)c+h.c.$ where the couplings are

$g_{\mathrm{dir}}=\frac{1}{2\hslash }\sum _{k=1,2}{E}_{\mathrm{jk}}{\varphi }_k^3{\varphi }_{ck}{{\xi }_k}^2{\xi }_k$ $g_s=\frac{1}{4\hslash }\sum _{k=1,2}{E}_{\mathrm{jk}}{\varphi }_{\mathrm{ck}}^2{\varphi }_k^2{\xi }_k^2$ $g_a=\frac{E_{J1}}{\hslash }{\varphi }_{a1}^2{\varphi }_{c1}{\varphi }_1{\xi }_1$ $g_b=\frac{E_{J2}}{\hslash }{\varphi }_{b2}^2{\varphi }_{c2}{\varphi }_2{\xi }_2$ $g_c=\frac{1}{2\hslash }\sum _{k=1,2}{E}_{\mathrm{jk}}{\varphi }_{\mathrm{ck}}^3{\varphi }_k{\xi }_k$

The remaining step now is to tune the second line of such that we obtain the term {circumflex over (Δ)}c+h.c. We can adjust the amplitude and the phase of ξ_1,2 such that|g_a|=|g_b|=ϵ_Δandarg(g_a)=arg(g_b)+π=0

For ϕ_a1≈ϕ_b2, ϕ_c1≈ϕ_c2, E_J1≈E_J2, and ϕ₁≈ϕ₂, the magnitude of the terms in the above equations for g_dir and g_c becomes minimal. The remaining g_dir can be canceled by setting ϵ_c=−g_dir, yielding

$\frac{H_{sys}}{\hslash }=-{\chi }_{ac}{a}^{†}{\mathrm{ac}}^{†}c-{\chi }_{\mathrm{bc}}{b}^{†}{\mathrm{bc}}^{†}c-\sum _{r=a,b,c}\frac{{\chi }_{\mathrm{rr}}}{2}{r}^{\mathrm{†2}}{r}^2-\left(g_{s}c+g_c{c}^{†}c+{ϵ}_{\Delta }\hat{\Delta }\right)c+h.c.$

Hence we can engineer the displacement term of the cavity mode c to be proportional to the error syndrome measurement operator{circumflex over (Δ)}=b^†b−a^†a

By applying this displacement to the cavity 540, the state of the cavity 540 may be engineering to be indicative of Δ. Accordingly, reading the state of the cavity 540 may indicate the value of Δ and an appropriate operation may be performed to correct for any error indicated by the state. An illustrative example of how such operations may be performed in practice is described in International Patent Application No. PCT/US2017/017534, filed Feb. 10, 2017, entitled “Techniques For Control Of Quantum Systems And Related Systems And Methods”, which is hereby incorporated by reference.

An alternative approach to correcting photon loss errors is shown in FIG. 9A. In system 900, portions of the system on the left are identical to those shown in FIG. 5 and discussed above. In particular, the above-described process for protecting against dephasing errors in the two-mode state of cavities 511 and 512 may be applied within system 900. However, system 900 presents an alternative approach to FIG. 5 in that the two Josephson junctions and additional cavity that may implement error syndrome detection are replaced by a Josephson junction 944 coupled to both cavities 511 and 512 via respective Superconducting Nonlinear Asymmetric Inductive eLements (SNAILs) 941 and 942. Elements 941, 942 and 944 provide an autonomous approach to correcting photon loss errors as discussed below through application of suitable drive waveforms 948 to the junction 944.

At a high level, the alternate approach provided by system 900 is based on compensating single photon losses in either cavity 511 or 512 by adding photon gain jump operators that are conditional on Δ=±1 through application of energy to the junction 944. The SNAILs 941 and 942 each comprise a superconducting loop of a number of comparatively large Josephson junctions and a one smaller Josephson junction with tunneling energies of E_J and αE_J, respectively. Further details of structure of a SNAIL are described in “3-wave mixing Josephson dipole element,” Appl. Phys. Lett. 110, 222603 (2017), which is incorporated herein by reference in its entirety.

For mode a, the jump operator isF(1)=a^†P_Δ=1 Similarly,F(−1)=b^†P_Δ=−1 for mode b. A difference between these jumps and the ideal continuous QEC jumps with Δ=±1 is the extra raising operator. Even though this recovery implements first-order dephasing errors aa^† or bb^† when combined with the preceding loss event, these jumps may have a negligible effect on the code states for sufficiently large γ.

We now sketch this proposal, focusing for the sake of example on correction for Δ=1 (mode a loss) only. In the exposition below, we apply perturbation theory sequentially. However, all terms could also be introduced simultaneously in a more involved calculation that yields the same low-order result with higher order corrections.

The SNAILs 941 and 942, which are three-wave mixers, provide Jaynes-Cummings type couplings between the junction and respective cavities 511 and 512 without any additional Kerr nonlinearities. We thus begin with a two-mode driven Jaynes-Cummings HamiltonianH=ω_aa^†a+ω_bb^†b+½ω_qσ_z +g(aσ₊e^−iωpat+bσ₊e^−iωpbt+h.c.)with cavity frequencies ω_a,b, Josephson junction frequency ω_g, real pump drive g, and pump frequencies ω_pa, ω_pb (where the drive 948 is also referred to as a pump drive). Setting ω_pa=ω_a−ω_q+δ and ω_pb=ω_b−ω_q−δ and going into the rotating frame with respect to H₀=ω_aa^†a+ω_bb^†b+½ω_qσ_z+δ{circumflex over (Δ)} for δ>0. The pump strengths are then set so we are in the dispersive regime: gk^†k<<δ with k^†k being the average occupation number in cavity k∈{a, b}.

In the dispersive limit, the Hamiltonian becomes

$\frac{g^2}{\delta }\hat{\Delta }{\sigma }_Z\equiv {\Omega }_{\Delta }\hat{\Delta }{\sigma }_Z$

For each pair of levels, the two cavities are in the subspace of fixed Δ while the junction is either in |g for the bottom level or in |e for the top.

In contrast to the approach described with respect to FIGS. 5, 6 and 7, the system of FIG. 9A does not perform a discrete readout of the error syndrome, but rather continuously performs the correction operation. As shown in FIG. 9B, the engineered cross-Kerr interaction described above induces a {circumflex over (Δ)}-dependent junction frequency. For each pair of levels, the cavities 511 and 512 are in the subspace of fixed Δ while the junction is either in |g for the bottom level or |e for the higher level.

When a photon loss event occurs in one of the cavities, the logical qubit stored in the two cavities is transferred into a different Δ subspace while the junction remains in |g. In the context of FIG. 9B, this shifts the subspace left or right amongst the different sets of junction energy levels shown for each value of Δ (with only a subset of the infinite possible values of Δ shown in the figure). The above Δ-dependent junction frequency is responsible for the “telescoping” of the junction energy levels in the schematic 950 of FIG. 9B.

Since the junction frequency depends on the error syndrome, we can add photons to the cavity selectively depending on {circumflex over (Δ)}. In particular, for Δ=1, the SNAILs may couple the junction to cavity a via the (counter-rotating) term at a^†σ₊ exp(−iω_CRt) with frequency ω_CR in the rotating frame with respect to H₀. We set ω_CR=3Ω_Δ, the frequency of the effective two-level system at Δ=1.

The Hamiltonian is thenH₁=Ω_Δ{circumflex over (Δ)}σ_z+g_Δ(a^†σ₊e^−3iΩΔt+h.c.)with g_Δ<<Ω_Δ<<δ. Going into the rotating frame with respect to Ω_Δ{circumflex over (Δ)}σ_z yields the desired transition:{tilde over (H)}=g_Δ(a^†σ₊e^{2iΩΔ({circumflex over (Δ)}-1)t}h.c.)

Furthermore, assuming the junction decays with jump operator √{square root over (Γσ₋)}, and adiabatically eliminating the junction, yields a dissipator with jump operator

$\frac{4g_{\Delta }^2}{\Gamma }F\left(1\right).$ Thus, we have our desired result to first-order.

Application of the above-described conditional energy pulse 948 will, conditional on the value of {circumflex over (Δ)}, drive the junction to its excited state |e whilst simultaneously applying an operation a, a^†, b, b^†, aa, etc. These energy pulses are shown in FIG. 9B using double-headed arrows. Subsequent to such a change in state, the junction may then decay from |e to |g to complete the process, with this decay shown as a wavy arrow in FIG. 9B. FIG. 9C includes schematic 960 which shows an example of this process when correcting a single loss event in cavity a. Upon the event, denoted by the thick arrow, the logical qubit stored in the two cavities is transferred into the subspace P_Δ=1 while the junction remains in |g. The applied pulse (double-headed arrow) then drives the junction to its excited state |e while simultaneously applying a^† to the cavity system and thereby returning the logical state back to P_Δ=0.

It will be appreciated that the approach of FIGS. 9A-9C is an example of the error syndrome detection circuit 240 shown in FIG. 2 in the sense that, whilst there is no explicit detection of the change in state due to photon loss, such detection is implicit in that operations are performed that are dependent upon there being a photon loss. As such, elements 941, 942 and 944 in FIG. 9A should be viewed as an example of error syndrome detection circuit 240 of FIG. 2.

FIG. 10 is a block diagram of a circuit quantum electrodynamics system, according to some embodiments. System 1000 includes a system 1001 in addition to electromagnetic radiation source 1030, controller 1040 and storage medium 1050. In some embodiments, a library of precomputed drive waveforms may be stored on a computer readable storage medium and accessed in order to apply said waveforms to system 200 as shown in FIG. 2. In the example of FIG. 10, controller 1040 may be configured to access drive waveforms 1052 stored on storage medium 1050 (e.g., in response to user input provided to the controller) and to control the electromagnetic radiation source 1030 to apply one or more drive waveforms ε₁ (t) to the mixing circuit 220. Optionally, one or more drive waveforms ε₂ (t) may be applied to the error syndrome detection circuit 240.

As non-limiting examples, ε₁ (t) may include drives which, as discussed above, protect against dephasing errors in the implementation of FIG. 5. That is, ε₁(t) may comprise drive waveforms 525. ε₂ (t) may be applied, in some embodiments, to cavity 540 in the implementation of FIG. 5 to detect (and in some cases, correct) photon error detected via the implementation of FIG. 5 as discussed above. That is, ε₂ (t) may comprise drive waveforms 545. In some embodiments, ε₂ (t) may be applied to Josephson junction 944 in the implementation of FIG. 9 to autonomously correct for photon losses. That is, ε₂ (t) may comprise drive waveforms 948. It will be appreciated that, in the example of FIG. 10, drive waveforms ε₁ (t) and ε₂ (t) may each include any number of energy pulses having any suitable frequencies and may be directed to the same, or different portions of mixing circuit 220 or error syndrome detection circuit 240.

Having thus described several aspects of at least one embodiment of this invention, it is to be appreciated that various alterations, modifications, and improvements will readily occur to those skilled in the art.

Such alterations, modifications, and improvements are intended to be part of this disclosure, and are intended to be within the spirit and scope of the invention. Further, though advantages of the present invention are indicated, it should be appreciated that not every embodiment of the technology described herein will include every described advantage. Some embodiments may not implement any features described as advantageous herein and in some instances one or more of the described features may be implemented to achieve further embodiments. Accordingly, the foregoing description and drawings are by way of example only.

The above-described embodiments of the technology described herein can be implemented in any of numerous ways. For example, some embodiments may be implemented, at least in part, using hardware, software or a combination thereof. When implemented at least in part in software, the software code can be executed on any suitable processor or collection of processors (e.g., controller 1040 in FIG. 10, whether provided in a single computer or distributed among multiple computers. Such processors may be implemented as integrated circuits, with one or more processors in an integrated circuit component, including commercially available integrated circuit components known in the art by names such as CPU chips, GPU chips, microprocessor, microcontroller, or co-processor. Alternatively, a processor may be implemented in custom circuitry, such as an ASIC, or semi-custom circuitry resulting from configuring a programmable logic device. As yet a further alternative, a processor may be a portion of a larger circuit or semiconductor device, whether commercially available, semi-custom or custom. As a specific example, some commercially available microprocessors have multiple cores such that one or a subset of those cores may constitute a processor. Though, a processor may be implemented using circuitry in any suitable format.

Various aspects of the present invention may be used alone, in combination, or in a variety of arrangements not specifically discussed in the embodiments described in the foregoing and is therefore not limited in its application to the details and arrangement of components set forth in the foregoing description or illustrated in the drawings. For example, aspects described in one embodiment may be combined in any manner with aspects described in other embodiments.

Also, the invention may be embodied as a method, of which an example has been provided. The acts performed as part of the method may be ordered in any suitable way. Accordingly, embodiments may be constructed in which acts are performed in an order different than illustrated, which may include performing some acts simultaneously, even though shown as sequential acts in illustrative embodiments.

Further, some actions are described as taken by a “user.” It should be appreciated that a “user” need not be a single individual, and that in some embodiments, actions attributable to a “user” may be performed by a team of individuals and/or an individual in combination with computer-assisted tools or other mechanisms.

Use of ordinal terms such as “first,” “second,” “third,” etc., in the claims to modify a claim element does not by itself connote any priority, precedence, or order of one claim element over another or the temporal order in which acts of a method are performed, but are used merely as labels to distinguish one claim element having a certain name from another element having a same name (but for use of the ordinal term) to distinguish the claim elements.

Also, the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of “including,” “comprising,” or “having,” “containing,” “involving,” and variations thereof herein, is meant to encompass the items listed thereafter and equivalents thereof as well as additional items.

Claims

1. A system comprising: a circuit quantum electrodynamics system that includes: a logical qubit comprising first and second quantum mechanical oscillators;a mixing circuit coupled to each of the first and second quantum mechanical oscillators; anda quantum error syndrome detection circuit coupled to each of the first and second quantum mechanical oscillators;at least one controller; andat least one computer readable medium storing instructions that, when executed by the at least one controller, perform a method comprising: applying, using the at least one controller, a plurality of drive waveforms to the mixing circuit, the plurality of drive waveforms having different frequencies; andconcurrent with said application of the plurality of drive waveforms, detecting, by the quantum error syndrome detection circuit, a change in energy state of the logical qubit.

2. The system of claim 1, wherein the plurality of drive waveforms are configured to drive a dissipation of energy from the first and second quantum mechanical oscillators to the mixing circuit.

3. The system of claim 1, wherein the mixing circuit is coupled to the first and second quantum mechanical oscillators via a first Josephson junction.

4. The system of claim 3, wherein the plurality of drive waveforms are configured to produce an exchange of energy between the first and second quantum mechanical oscillators and the first Josephson junction.

5. The system of claim 4, wherein said exchange of energy is configured to passively protect against dephasing errors of the logical qubit.

6. The system of claim 1, wherein the first and second quantum mechanical oscillators exhibit different resonant frequencies.

7. The system of claim 1, wherein the first and second quantum mechanical oscillators each comprise a microwave cavity.

8. The system of claim 1, wherein the quantum error syndrome detection circuit comprises a fourth quantum mechanical oscillator independently coupled to each of the first and second quantum mechanical oscillators.

9. The system of claim 8, wherein the quantum error syndrome detection circuit further comprises second and third Josephson junctions, wherein the fourth quantum mechanical oscillator is coupled to the first quantum mechanical oscillator via the second Josephson junction, and wherein the fourth quantum mechanical oscillator is coupled to the second quantum mechanical oscillator via the third Josephson junction.

10. The system of claim 9, wherein the instructions are further configured to, when executed by the at least one controller, apply first and second pumps to the second and third Josephson junctions, respectively.

11. The system of claim 10, wherein the instructions are further configured to, when executed by the at least one controller, detect an energy displacement of the fourth quantum mechanical oscillator caused by application of the first and second pumps, wherein said energy displacement is indicative of the change in the energy state of the logical qubit.

12. The system of claim 1, wherein the quantum error syndrome detection circuit comprises a fourth Josephson junction coupled to each of the first and second quantum mechanical oscillators.

13. The system of claim 12, wherein the change in the energy state of the logical qubit is implicitly detected via application of one or more drive waveforms to the first quantum mechanical oscillator that produce a change in the energy state of the first quantum mechanical oscillator which is dependent upon a state of the fourth Josephson junction.

14. A method of performing quantum error correction within a circuit quantum electrodynamics system that includes a logical qubit comprising first and second quantum mechanical oscillators, a mixing circuit coupled to each of the first and second quantum mechanical oscillators; and quantum error syndrome detection circuit coupled to each of the first and second quantum mechanical oscillators, the method comprising: applying, using at least one controller, a plurality of drive waveforms to the mixing circuit, the plurality of drive waveforms having different frequencies; andconcurrent with said application of the plurality of drive waveforms, detecting, by the quantum error syndrome detection circuit, a change in energy state of the logical qubit.

15. The method of claim 14, wherein the plurality of drive waveforms drive a dissipation of energy from the first and second quantum mechanical oscillators to the mixing circuit.

16. The method of claim 14, wherein the plurality of drive waveforms produce an exchange of energy between the first and second quantum mechanical oscillators and a first Josephson junction, and wherein the mixing circuit is coupled to the first and second quantum mechanical oscillators via the first Josephson junction.

17. A circuit quantum electrodynamics system comprising: a physical qubit;a logical qubit comprising a first bosonic system and a second bosonic system, the first and second bosonic systems being coupled to the physical qubit;a controller configured to encode, during operation of the circuit quantum electrodynamics system, a state of the physical qubit in a state of the logical qubit, the encoded state being a multi-component coherent state of the first and second bosonic systems; andan error syndrome detector coupled to the logical qubit and configured to detect an error in said state of the logical qubit based on one or more quantum non-destructive measurements of a boson number difference between the first and second bosonic systems.

18. The circuit quantum electrodynamics system of claim 17, wherein the error syndrome detector comprises a third bosonic system independently coupled to each of the first and second bosonic systems.

19. The circuit quantum electrodynamics system of claim 18, wherein the error syndrome detector further comprises second and third Josephson junctions, wherein the third bosonic system is coupled to the first bosonic system via the second Josephson junction, and wherein the third bosonic system is coupled to the second bosonic system via the third Josephson junction.

20. The circuit quantum electrodynamics system of claim 18, wherein the error syndrome detector is further configured to detect an energy displacement of the third bosonic system caused by application of energy to the third bosonic system, wherein said energy displacement is indicative of the boson number difference between the first and second bosonic systems.