Unconditional measurement-based quantum computation with optomechanical continuous variables

Unconditional measurement-based quantum computation
with optomechanical continuous variables

Oussama Houhou o.houhou@qub.ac.uk School of Mathematics and Physics, Queen’s University Belfast, BT7 1NN, UK Laboratory of Physics of Experimental Techniques and Applications, University of Médéa, Médéa 26000, Algeria    Darren W. Moore dmoore32@qub.ac.uk Department of Optics, Palacký University, 17. listopadu 1192/12, 771 46 Olomouc, Czech Republic School of Mathematics and Physics, Queen’s University Belfast, BT7 1NN, UK    Sougato Bose s.bose@ucl.ac.uk Department of Physics and Astronomy, University College London, London WC1E 6BT, UK    Alessandro Ferraro a.ferraro@qub.ac.uk School of Mathematics and Physics, Queen’s University Belfast, BT7 1NN, UK o.houhou@qub.ac.uk School of Mathematics and Physics, Queen’s University Belfast, BT7 1NN, UK Laboratory of Physics of Experimental Techniques and Applications, University of Médéa, Médéa 26000, Algeria dmoore32@qub.ac.uk School of Mathematics and Physics, Queen’s University Belfast, BT7 1NN, UK s.bose@ucl.ac.uk Department of Physics and Astronomy, University College London, London WC1E 6BT, UK a.ferraro@qub.ac.uk School of Mathematics and Physics, Queen’s University Belfast, BT7 1NN, UK

Universal quantum computation encoded over continuous variables can be achieved via Gaussian measurements acting on entangled non-Gaussian states. However, due to the weakness of available nonlinearities, generally these states can only be prepared conditionally, potentially with low probability. Here we show how universal quantum computation could be implemented unconditionally using an integrated platform able to sustain both linear and quadratic optomechanical-like interactions. Specifically, considering cavity opto- and electro-mechanical systems, we propose a realisation of a driven-dissipative dynamics that deterministically prepares the required non-Gaussian cluster states — entangled squeezed states of multiple mechanical oscillators suitably interspersed with cubic-phase states. We next demonstrate how arbitrary Gaussian measurements on the cluster nodes can be performed by continuously monitoring the output cavity field. Finally, the feasibility requirements of this approach are analysed in detail, suggesting that its building blocks are within reach of current technology.

Introduction.— Measurement-based quantum computation (MBQC) is a powerful approach to process information encoded in quantum systems Briegel et al. (2009), which requires solely local measurements on an entangled state (cluster state) Raussendorf and Briegel (2001); Raussendorf et al. (2003). This approach gives significant theoretical insights into fundamental questions about the origin of the power of quantum computing Van den Nest et al. (2006); Gross et al. (2009); Bremner et al. (2009); Anders and Browne (2009); Raussendorf (2013); Bermejo-Vega et al. (2017), and it offers promising applicative opportunities provided large enough clusters can be built, including the demonstration of quantum computational supremacy Hangleiter et al. (2018); Bermejo-Vega et al. (2018) and the realisation, in condensed matter systems Brennen and Miyake (2008); Cai et al. (2010); Li et al. (2011); Wei et al. (2011); Aolita et al. (2011); Else et al. (2012); Wei and Raussendorf (2015); Miller and Miyake (2015), of fault-tolerant processors with high resilience thresholds Raussendorf and Harrington (2007); Raussendorf et al. (2007).

In view of the relevance of MBQC, major efforts have been devoted to its experimental implementation. In the setting of finite-dimensional (discrete-variable) quantum systems, various experimental demonstrations of small-size MBQC have been reported Walther et al. (2005); Kiesel et al. (2005); Prevedel et al. (2007); Tame et al. (2007); Lu et al. (2007); Vallone et al. (2008); Barz et al. (2012); Bell et al. (2014); Lanyon et al. (2013). However, the largest clusters to date have been generated in the context of continuous-variable (CV) systems Braunstein and Van Loock (2005); Serafini (2017), with photonic clusters composed of up to one million modes Yokoyama et al. (2013); Chen et al. (2014); Roslund et al. (2014); ichi Yoshikawa et al. (2016); Cai et al. (2017); Rigas et al. (2012). Such achievements stem from the fact that these clusters belong to the class of Gaussian states Ferraro et al. (2005); Weedbrook et al. (2012); Adesso et al. (2014); Genoni et al. (2016), and CV Gaussian entanglement is generally available unconditionally (deterministically), contrary to discrete-variable systems whose entanglement typically relies on post-selection 111See Refs. Lindner and Rudolph (2009); Schwartz et al. (2016); Pichler et al. (2017) for progresses towards deterministic generation of discrete variable clusters and Refs. Spiller et al. (2006); Anders et al. (2010); Roncaglia et al. (2011); Proctor et al. (2017); Gallagher and Ferraro (2018) for alternative measurement-based approaches.. Despite such remarkable progress, in order to realize universal computation Lloyd and Braunstein (1999), these photonic clusters have to be either equipped with non-Gaussian measurements Menicucci et al. (2006); Gu et al. (2009) or interspersed with non-Gaussian states Menicucci (2014). Unfortunately, both strategies require high order non-linearities, which are hard to implement deterministically in optics and in fact stand as a major roadblock 222Notice that much effort has been devoted to counteract this issue, leading to proposals in which the necessary non-Gaussian elements can be obtained on-demand Marek et al. (2011); Yukawa et al. (2013); Marshall et al. (2015); Miyata et al. (2016); Marek et al. (2017); Takeda and Furusawa (2017); however these still require the use of quantum memories which are hard to realize Lvovsky et al. (2009); Yoshikawa et al. (2013) and that we avoid here.. As a remedy, we propose here to use non-linear optomechanical systems, with the aim of providing a feasible path to unlock the full potential of unconditional MBQC.

Our approach is motivated by recent experimental breakthroughs in cavity optomechanics Milburn and Woolley (2011); Aspelmeyer et al. (2014), which lends itself as a disruptive new platform for CVs in which the information carrier is embodied in the centre of mass motion of a mechanical oscillator. Indeed ground state cooling Teufel et al. (2011); O’Connell et al. (2010); Noguchi et al. (2016a), squeezing beyond the parametric limit Lei et al. (2016); Wollman et al. (2015); Pirkkalainen et al. (2015), two-oscillator entanglement Ockeloen-Korppi et al. (2018); Riedinger et al. (2018) and non-locality Marinkovic et al. (2018) have been achieved experimentally, with further scalability and integrability within reach Massel et al. (2012); Damskägg et al. (2016); Grass et al. (2016); Nielsen et al. (2017); Noguchi et al. (2016b). Crucially, optomechanics has a significant advantage to photonics in the unconditional non-linearity embedded in the radiation pressure dynamics Bhattacharya et al. (2008); Thompson et al. (2008). For driven systems this manifests primarily as a quadratic coupling in the position of the oscillator Thompson et al. (2008); Woolley et al. (2008); Hertzberg et al. (2010); Rocheleau et al. (2010); Nunnenkamp et al. (2010); Purdy et al. (2010); Sankey et al. (2010); Hill et al. (2011); Karuza et al. (2013); Flowers-Jacobs et al. (2012); Massel et al. (2012); Li et al. (2012); Hill (2013); Doolin et al. (2014); Kaviani et al. (2015); Paraïso et al. (2015); Lee et al. (2015); Kim et al. (2015); Abdi and Hartmann (2015); Kalaee et al. (2016); Fonseca et al. (2016); Brawley et al. (2016); Leijssen et al. (2017); Zhang et al. (2017); Dellantonio et al. (2018).

Here we consider a driven-dissipative opto-mechanical system. By taking advantage of the control over the mechanical state granted by externally driving the cavity, and arranging for either dissipative engineering Kronwald et al. (2013); Ikeda and Yamamoto (2013); Houhou et al. (2015) or continuous monitoring Clerk et al. (2008); Genoni et al. (2016); Moore et al. (2017), we are able to provide schemes for the deterministic preparation of non-Gaussian cluster states and local measurements sufficient to achieve computational universality. The integration of these schemes into a single experimental platform constitutes, as far as we know, the first proposal for universal MBQC with CVs that can be implemented unconditionally.

Measurement based computation with CVs.— As said, MBQC is predicated on the existence of a highly entangled multipartite resource state known as the cluster state. For our purposes, a cluster state is associated with a mathematical lattice graph of vertices , and edges that define the adjacency matrix with entries if (with ). Consider an -oscillator system, with each oscillator characterised by the canonical position and momentum operators, being their respective annihilation operator. The CV cluster state Zhang and Braunstein (2006); Menicucci et al. (2006) is operationally defined by first preparing all vertices (embodied by the oscillators) in a product state of momentum-squeezed vacua , where , , , and is a shorthand for the degree of squeezing. Then, controlled-phase operations are applied for any edge . These can be compactly written defining the multi-oscillator operator , with . Consequently, the resulting standard cluster state is given by (see Fig. 1); this is a Gaussian state and the degree of squeezing (with ) determines its quality for computational purposes Gu et al. (2009).

Figure 1: Circuit representing the preparation of a cluster state with squeezing and adjacency matrix . In the absence (presence) of cubic operations —given in the dashed box— the standard (non-Gaussian) cluster is obtained.

The computation proceeds via a series of local projective measurements on the cluster nodes. These measurements implement the gates of the program to be computed, whose output is embodied in the state of the non-measured nodes. The Lloyd-Braunstein criterion Lloyd and Braunstein (1999), first developed for circuit-based computation, allows a distinction to be drawn between Gaussian and non-Gaussian gates. A finite set of Gaussian gates is sufficient to perform any multimode Gaussian operation. However, it is only when an additional non-Gaussian gate is at disposal that universality is unlocked, in the sense that any Hamiltonian can be simulated to arbitrary precision. In MBQC, Gaussian measurements on the cluster are sufficient to implement arbitrary Gaussian gates Gu et al. (2009), including in extremely compact ways Ferrini et al. (2013). On the other hand, as mentioned, several proposals for implementing non-Gaussian gates are extant in the literature Ghose and Sanders (2007); Gu et al. (2009); Marek et al. (2011); Yukawa et al. (2013); Marshall et al. (2015); Miyata et al. (2016); Marek et al. (2017); Takeda and Furusawa (2017). Here we focus on a method in which the standard cluster is modified using non-Gaussian resources — called cubic-phase states Gottesman et al. (2001). This modified non-Gaussian cluster is particularly advantageous for scaling to large numbers of operations since it allows for the measurement strategy to remain Gaussian Gu et al. (2009); Gottesman et al. (2001).

We will first present a general exposition of the optomechanics model we wish to base our proposal on, and then introduce two complementary schemes allowing us to prepare the modified non-Gaussian cluster and perform on it arbitrary Gaussian measurements.

Optomechanics implementation.— Consider an array of mechanical resonators, each with distinct frequency , immersed in a cavity field of frequency and driven by an external field . The Hamiltonian for such a system is . Due to radiation pressure the cavity frequency becomes dependent on the mechanical positions Aspelmeyer et al. (2014). Expanding in powers of these latter, we write: , with and . In addition, we consider the case of a multi-tone drive, with the complex driving amplitudes and the driving frequencies. The standard linearisation procedure for an externally driven cavity Aspelmeyer et al. (2014) yields, in the frame rotating with the free terms, the following Hamiltonian PRL ()


where the are the detunings of the field with the cavity, and are the amplifications of the single phonon-photon couplings due to the external driving. Then, we consider four driving fields per each mechanical resonator with detunings and amplitudes (). Moreover, we consider an additional drive that is resonant with the cavity (), with amplitude . Hamiltonian (1), in the rotating wave approximation (RWA), becomes PRL ()


with , (; ) and . Notice that independent control over each term in the Hamiltonian (2) is possible 333Note that the parameters , , can be tuned by varying both the bare optomechanical couplings and the driving amplitudes, while is set by changing only., which is in turn crucial for our purposes. The aforementioned RWA holds in a regime satisfying and (, , ), given that the frequencies do not overlap PRL ().

As said, dissipation is central for our aims. We model the evolution of the system by a master equation in which the cavity mode dissipates at a rate and the mechanical oscillators are in contact with a thermal bath Carmichael (1993); Gardiner and Zoller (2000):


where and denote the mechanical damping rate and mean phonon number for the -th mechanical oscillator. The standard super-operator for Markovian dissipation is denoted as ().

The cubic phase state.— The finitely-squeezed cubic phase state of a single system is defined as Gottesman et al. (2001)


A core result of our proposal is that the cubic phase state of a single mechanical oscillator can be unconditionally generated as the steady state of the dynamics given in Eq. (3) (with ), applying suitable drive amplitudes and phases. The coefficients of the linear terms, and , are associated only with Gaussian steady states Houhou et al. (2015). Indeed, the ratio of the amplitudes of these determines the degree of squeezing 444We describe the level of squeezing as  dB Gu et al. (2009). of the steady state Houhou et al. (2015); Kronwald et al. (2013). Non-Gaussianity at the steady state derives instead from the remaining coefficients as follows. By choosing the driving strengths as , with , we obtain the Hamiltonian


It can be proven analytically PRL () that, neglecting the mechanical thermal noise, the master equation (3) has the steady state where is the vacuum state of the cavity and is the mechanical finitely squeezed cubic phase state defined in Eq. (4) with . The stability condition of the system’s dynamics is inherited from the linear system: . Notice that for this Hamiltonian, the cubic phase state is the only steady state regardless of the initial state PRL ().

In order to consider the effect of non-zero mechanical noise, we numerically find the steady state of Eq. (3) and then we calculate the fidelity between the latter and the state in Eq. (4). This is shown in Fig. 2 where we plot the fidelity as a function of the mean phonon number of the bath and the mechanical damping rate. As expected, the mechanical noise has a noxious effect on the target cubic phase state; the higher the temperature (quantified by ) or mechanical damping rate () the lower the fidelity.

Figure 2: Fidelity of the noisy cubic phase state with the noiseless one as a function of the mean phonon number () and mechanical damping rate (). Each point of the plot was obtained with the following parameters: (), , and .

This analysis shows that a cubic phase state can be generated in the massive mechanical oscillator of an optomechanics experiment. This is a result of interest in its own, given the highly non classical character of such a state — which displays a non-positive Wigner function and a high degree of quantum non-Gaussianity Takagi and Zhuang (2018); Albarelli et al. (2018) — and its deterministic attainability. As mentioned, for this state to be considered as a resource for computing, we must also show that it can be embedded in a standard Gaussian cluster state 555A seminal proposal to realize, via optomechanical-like interactions, states potentially useful for MBQC is given in Ref. Pirandola et al. (2006); however, there a non-linearized and probabilistic approach is considered. We also notice that the recent deterministic proposal in Ref. Brunelli et al. (2018) use similar techniques to the ones presented here but the states obtained there are not useful for MBQC (see also Brunelli and Houhou (2018) for further analysis)..

Non-Gaussian cluster states.— We aim to generate a modified non-Gaussian cluster state sufficient to perform universal computation by interspersing the standard state with cubic-phase states. In particular we will now show how the dissipative dynamics described by Eq. (3) can be adapted to generate the state , where denotes the cubic non-linearities, and we have defined and (see Fig. 1). The state allows the implementation of universal computation since it can be composed of nodes with zero non-linearity, as the standard Gaussian one, and nodes with . For any given computation, Gaussian measurements will then “tailor” this non-Gaussian cluster accordingly to the program to be implemented. In this way, cubic gates can be implemented only when needed 666Similar tailoring techniques have been considered in the context of both continuous Menicucci (2014) and discrete variables Bermejo-Vega et al. (2017). In the latter, Gaussian and non-Gaussian gates are substituted with Clifford and non-Clifford ones..

Adapting the Hamiltonian switching scheme considered in Refs. Ikeda and Yamamoto (2013); Houhou et al. (2015), one can generate the state via dissipation engineering. The switching scheme involves steps such that at each one the driving fields are tuned to implement the transformation . This implies that, at the step, the Hamiltonian is where is a positive parameter. At each step, the system is allowed to reach its steady state (i.e., the vacuum of the collective mode ) and then the Hamiltonian is switched, by modifying the driving fields, for the next step to begin (see Ref. PRL () for details). Therefore, if the system is initially in vacuum (and neglecting the mechanical damping), after the steps the mechanical state is given by the target cluster state, in the basis of the local modes 777We should mention that our switching scheme introduced here is not only a generalisation of a previous protocol Houhou et al. (2015) for the generation of Gaussian cluster states, but also conforms with the canonical preparation of Gaussian cluster states if we restrict our selves to first sideband drivings only..

Fig. 3 demonstrates the effectiveness of the switching scheme for generating a two-node non-Gaussian cluster. In the absence of mechanical noise (solid red line), the fidelity with the target state increases monotonically in each step and it reaches unit fidelity at the steady state (at the end of the second step, provided longer evolution time is allowed). When the mechanical environment is considered (dot dashed line), the fidelity reaches a maximum (during the second step) before the noise starts to negatively affect the quality of the target cluster state. As already seen in Fig. 2, the thermal noise has a detrimental effect on the performance of the switching scheme, however high degrees of fidelities can still be achieved. Part of this negative effect is due to the fact that the oscillators are assumed to be initialized in thermal equilibrium with their environment (with mean phonon numbers and and mechanical damping ), rather than in the ground state. This effect can then be circumvented to a large degree by first independently cooling the oscillators (red detuned sideband cooling) Marquardt et al. (2007); Wilson-Rae et al. (2007). This can be seen in the dashed blue curve of Fig. 3, which in fact closely approximates the noiseless scenario.

Figure 3: The fidelity of the preparation of a two-node non-Gaussian cluster state. The nodes of the cluster consist of a squeezed state and a cubic phase state with same amount of squeezing. We used the following parameters: (), , , and evolution duration . Pre-cooling (dashed line) the oscillators close to the ground state greatly increases the (maximum) achievable fidelity of the scheme.

Local Gaussian measurements— For the non-Gaussian cluster state above to be useful in quantum computation, one finally requires the capacity to perform Gaussian measurements on individual nodes. Unfortunately, the mechanical modes embodying our cluster state are not directly accessible to measurement and must be probed instead using the cavity field as a detector. Conveniently, since the resonators are assumed to have distinct and well-spaced frequencies PRL (), we may address each oscillator individually. In particular, by properly driving the system, it is possible to engineer a quantum non-demolition (QND) interaction between the cavity position quadrature and an arbitrary quadrature of any given oscillator Clerk et al. (2008); Moore et al. (2017).

Consider again Hamiltonian (2) with (), i.e. addressing only the first sidebands. Let and . In this case one has a sum of QND interactions, , where is the cavity position quadrature and is an arbitrary quadrature of the mechanical mode . Each oscillator can be addressed in turn by setting all but the amplitude of interest to zero. In this case we have an -step process with each step described by . Let us stress that the mechanical quadrature to be measured, which in turn depends on the program to be implemented, is simply selected by the phase of the external driving. Continuously monitoring, via homodyne detection, the output cavity field’s position quadrature drives the mechanical system towards an eigenstate of the chosen quadrature (represented by a vacuum state squeezed along an appropriate axis defined by ). For the purposes of computation, this is equivalent to performing a projective quadrature measurement directly onto the cluster state Moore et al. (2017). As said, the latter are in turn sufficient to perform any multimode operation, when operating on the non-Gaussian cluster .

In PRL () we provide an example of how to implement the minimal building block of universal MBQC by using the tools introduced so far. In particular, we consider the universal non-Gaussian gate defined as the operator Weedbrook et al. (2012) — called cubic phase gate — and show that it can be reliably implemented on a squeezed state via local Gaussian measurements on the two-node non-Gaussian cluster of Fig. 3.

Experimental feasibility.— The protocol proposed above to prepare non-Gaussian cluster states requires physical platforms exhibiting linear and quadratic position coupling with the cavity field. Moreover, the system needs to operate in the resolved sideband regime and the conditions and (, , ) must be met to ensure the validity of the RWA used in our derivation of the dynamics. These requirements may be realised in current and near future experiments. In fact, there are many platforms that can be used to implement our scheme, including membrane-in-the-middle configurations Thompson et al. (2008); Sankey et al. (2010); Flowers-Jacobs et al. (2012); Karuza et al. (2013); Lee et al. (2015), ultracold atoms inside a cavity Purdy et al. (2010), photonic crystals Paraïso et al. (2015); Kalaee et al. (2016); Leijssen et al. (2017); Kaviani et al. (2015), circuit-QED Kim et al. (2015), electro-mechanical systems Woolley et al. (2008); Hertzberg et al. (2010); Rocheleau et al. (2010); Massel et al. (2012); Dellantonio et al. (2018), micro-disks Li et al. (2012); Hill et al. (2011); Hill (2013); Doolin et al. (2014), and optically levitated particles Chang et al. (2010); Abdi and Hartmann (2015); Fonseca et al. (2016). In particular, very large quadratic couplings are within reach of current experiments Paraïso et al. (2015); Kalaee et al. (2016); Hill et al. (2011). Also we mention that linear to quadratic ratios of up to may be obtained Zhang et al. (2017); Brawley et al. (2016).

Furthermore, the linear to quadratic couplings ratio can be improved by optimising the experimental design. For instance, one may exploit the membrane tilting in membrane-in-the-middle setups Thompson et al. (2008); Sankey et al. (2010) or fine positioning the microdisc in microtorid optomechanical systems Li et al. (2012). Also, our protocols can be implemented in electrical circuits by controlling the bias flux and coupling capacitance as proposed in Kim et al. (2015), or considering magnetically or optically levitated particles as suggested in Pino et al. (2018); Chang et al. (2010).

Conclusions and outlook.— Continuous-variable systems are convenient for fault-tolerant computation since they naturally offer high-dimensional spaces in which the discrete units of quantum information can be resiliently encoded Chuang et al. (1997); Gottesman et al. (2001); Lund et al. (2008); Menicucci (2014); Michael et al. (2016); Fukui et al. (2017), as recently proven experimentally in the context of circuit-based quantum computation Ofek et al. (2016); Rosenblum et al. (2018); Flühmann et al. (2018). In this respect, the alternative measurement-based approach considered here is promising, thanks to the availability of high threshold schemes Raussendorf and Harrington (2007); Raussendorf et al. (2007). In particular, we have shown that a setting where mechanical oscillators act as the information carriers, rather than photons, provides the advantage that the core ingredients for universal computation —non-Gaussian cluster states and Gaussian operations— can be realized unconditionally. This opens the way to deterministic fault-tolerant quantum computation in integrable platforms where linear and quadratic optomechanics-like interactions can be simultaneously achieved.

We thank M. Brunelli, M. Paternostro, and M. Sillanpää for helpful discussions. A.F. and O.H. acknowledge support from the EPSRC project EP/P00282X/1 and the EU Horizon2020 Collaborative Project TEQ (grant agreement nr. 766900). O.H. acknowledges support from the SFI-DfE Investigator programme (grant 15/IA/2864). D.M. acknowledges the Coordinator Support funds from Queen’s University Belfast.


  • Briegel et al. (2009) Hans J Briegel, David E Browne, Wolfgang Dür, Robert Raussendorf,  and Maarten Van den Nest, “Measurement-based quantum computation,” Nature Physics 5, 19 (2009).
  • Raussendorf and Briegel (2001) Robert Raussendorf and Hans J. Briegel, “A one-way quantum computer,” Phys. Rev. Lett. 86, 5188–5191 (2001).
  • Raussendorf et al. (2003) Robert Raussendorf, Daniel E Browne,  and Hans J Briegel, “Measurement-based quantum computation on cluster states,” Phys. Rev. A 68, 022312 (2003).
  • Van den Nest et al. (2006) Maarten Van den Nest, Akimasa Miyake, Wolfgang Dür,  and Hans J. Briegel, “Universal resources for measurement-based quantum computation,” Phys. Rev. Lett. 97, 150504 (2006).
  • Gross et al. (2009) D. Gross, S. T. Flammia,  and J. Eisert, “Most quantum states are too entangled to be useful as computational resources,” Phys. Rev. Lett. 102, 190501 (2009).
  • Bremner et al. (2009) Michael J. Bremner, Caterina Mora,  and Andreas Winter, “Are random pure states useful for quantum computation?” Phys. Rev. Lett. 102, 190502 (2009).
  • Anders and Browne (2009) Janet Anders and Dan E. Browne, “Computational power of correlations,” Phys. Rev. Lett. 102, 050502 (2009).
  • Raussendorf (2013) Robert Raussendorf, “Contextuality in measurement-based quantum computation,” Phys. Rev. A 88, 022322 (2013).
  • Bermejo-Vega et al. (2017) Juan Bermejo-Vega, Nicolas Delfosse, Dan E. Browne, Cihan Okay,  and Robert Raussendorf, “Contextuality as a resource for models of quantum computation with qubits,” Phys. Rev. Lett. 119, 120505 (2017).
  • Hangleiter et al. (2018) Dominik Hangleiter, Juan Bermejo-Vega, Martin Schwarz,  and Jens Eisert, “Anticoncentration theorems for schemes showing a quantum speedup,” Quantum 2, 65 (2018).
  • Bermejo-Vega et al. (2018) Juan Bermejo-Vega, Dominik Hangleiter, Martin Schwarz, Robert Raussendorf,  and Jens Eisert, “Architectures for quantum simulation showing a quantum speedup,” Phys. Rev. X 8, 021010 (2018).
  • Brennen and Miyake (2008) Gavin K. Brennen and Akimasa Miyake, “Measurement-based quantum computer in the gapped ground state of a two-body hamiltonian,” Phys. Rev. Lett. 101, 010502 (2008).
  • Cai et al. (2010) Jianming Cai, Akimasa Miyake, Wolfgang Dür,  and Hans J. Briegel, “Universal quantum computer from a quantum magnet,” Phys. Rev. A 82, 052309 (2010).
  • Li et al. (2011) Ying Li, Daniel E. Browne, Leong Chuan Kwek, Robert Raussendorf,  and Tzu-Chieh Wei, “Thermal states as universal resources for quantum computation with always-on interactions,” Phys. Rev. Lett. 107, 060501 (2011).
  • Wei et al. (2011) Tzu-Chieh Wei, Ian Affleck,  and Robert Raussendorf, “Affleck-kennedy-lieb-tasaki state on a honeycomb lattice is a universal quantum computational resource,” Phys. Rev. Lett. 106, 070501 (2011).
  • Aolita et al. (2011) Leandro Aolita, Augusto J Roncaglia, Alessandro Ferraro,  and Antonio Acín, “Gapped two-body hamiltonian for continuous-variable quantum computation,” Phys. Rev. Lett. 106, 090501 (2011).
  • Else et al. (2012) Dominic V. Else, Ilai Schwarz, Stephen D. Bartlett,  and Andrew C. Doherty, “Symmetry-protected phases for measurement-based quantum computation,” Phys. Rev. Lett. 108, 240505 (2012).
  • Wei and Raussendorf (2015) Tzu-Chieh Wei and Robert Raussendorf, “Universal measurement-based quantum computation with spin-2 affleck-kennedy-lieb-tasaki states,” Phys. Rev. A 92, 012310 (2015).
  • Miller and Miyake (2015) Jacob Miller and Akimasa Miyake, “Resource quality of a symmetry-protected topologically ordered phase for quantum computation,” Phys. Rev. Lett. 114, 120506 (2015).
  • Raussendorf and Harrington (2007) Robert Raussendorf and Jim Harrington, “Fault-tolerant quantum computation with high threshold in two dimensions,” Phys. Rev. Lett. 98, 190504 (2007).
  • Raussendorf et al. (2007) R Raussendorf, J Harrington,  and K Goyal, “Topological fault-tolerance in cluster state quantum computation,” New Journal of Physics 9, 199 (2007).
  • Walther et al. (2005) Philip Walther, Kevin J Resch, Terry Rudolph, Emmanuel Schenck, Harald Weinfurter, Vlatko Vedral, Markus Aspelmeyer,  and Anton Zeilinger, “Experimental one-way quantum computing,” Nature 434, 169 (2005).
  • Kiesel et al. (2005) Nikolai Kiesel, Christian Schmid, Ulrich Weber, Géza Tóth, Otfried Gühne, Rupert Ursin,  and Harald Weinfurter, “Experimental analysis of a four-qubit photon cluster state,” Phys. Rev. Lett. 95, 210502 (2005).
  • Prevedel et al. (2007) Robert Prevedel, Philip Walther, Felix Tiefenbacher, Pascal Böhi, Rainer Kaltenbaek, Thomas Jennewein,  and Anton Zeilinger, “High-speed linear optics quantum computing using active feed-forward,” Nature 445, 65 (2007).
  • Tame et al. (2007) M. S. Tame, R. Prevedel, M. Paternostro, P. Böhi, M. S. Kim,  and A. Zeilinger, “Experimental realization of deutsch’s algorithm in a one-way quantum computer,” Phys. Rev. Lett. 98, 140501 (2007).
  • Lu et al. (2007) Chao-Yang Lu, Xiao-Qi Zhou, Otfried Gühne, Wei-Bo Gao, Jin Zhang, Zhen-Sheng Yuan, Alexander Goebel, Tao Yang,  and Jian-Wei Pan, “Experimental entanglement of six photons in graph states,” Nature physics 3, 91 (2007).
  • Vallone et al. (2008) Giuseppe Vallone, Enrico Pomarico, Francesco De Martini,  and Paolo Mataloni, ‘‘Active one-way quantum computation with two-photon four-qubit cluster states,” Phys. Rev. Lett. 100, 160502 (2008).
  • Barz et al. (2012) Stefanie Barz, Elham Kashefi, Anne Broadbent, Joseph F. Fitzsimons, Anton Zeilinger,  and Philip Walther, “Demonstration of blind quantum computing,” Science 335, 303–308 (2012)http://science.sciencemag.org/content/335/6066/303.full.pdf .
  • Bell et al. (2014) BA Bell, DA Herrera-Martí, MS Tame, D Markham, WJ Wadsworth,  and JG Rarity, “Experimental demonstration of a graph state quantum error-correction code,” Nature communications 5, 3658 (2014).
  • Lanyon et al. (2013) B. P. Lanyon, P. Jurcevic, M. Zwerger, C. Hempel, E. A. Martinez, W. Dür, H. J. Briegel, R. Blatt,  and C. F. Roos, “Measurement-based quantum computation with trapped ions,” Phys. Rev. Lett. 111, 210501 (2013).
  • Braunstein and Van Loock (2005) Samuel L Braunstein and Peter Van Loock, “Quantum information with continuous variables,” Rev. Mod. Phys. 77, 513 (2005).
  • Serafini (2017) Alessio Serafini, Quantum Continuous Variables: A Primer of Theoretical Methods (CRC Press, 2017).
  • Yokoyama et al. (2013) Shota Yokoyama, Ryuji Ukai, Seiji C Armstrong, Chanond Sornphiphatphong, Toshiyuki Kaji, Shigenari Suzuki, Jun-ichi Yoshikawa, Hidehiro Yonezawa, Nicolas C Menicucci,  and Akira Furusawa, “Ultra-large-scale continuous-variable cluster states multiplexed in the time domain,” Nature Photonics 7, 982–986 (2013).
  • Chen et al. (2014) Moran Chen, Nicolas C Menicucci,  and Olivier Pfister, “Experimental realization of multipartite entanglement of 60 modes of a quantum optical frequency comb,” Phys. Rev. Lett. 112, 120505 (2014).
  • Roslund et al. (2014) Jonathan Roslund, Renné Medeiros De Araujo, Shifeng Jiang, Claude Fabre,  and Nicolas Treps, “Wavelength-multiplexed quantum networks with ultrafast frequency combs,” Nature Photonics 8, 109–112 (2014).
  • ichi Yoshikawa et al. (2016) Jun ichi Yoshikawa, Shota Yokoyama, Toshiyuki Kaji, Chanond Sornphiphatphong, Yu Shiozawa, Kenzo Makino,  and Akira Furusawa, “Invited article: Generation of one-million-mode continuous-variable cluster state by unlimited time-domain multiplexing,” APL Photonics 1, 060801 (2016).
  • Cai et al. (2017) Y Cai, J Roslund, G Ferrini, F Arzani, X Xu, C Fabre,  and N Treps, ‘‘Multimode entanglement in reconfigurable graph states using optical frequency combs,” Nature communications 8, 15645 (2017).
  • Rigas et al. (2012) Ioannes Rigas, Christian Gabriel, Stefan Berg-Johansen, Andrea Aiello, Peter van Loock, Ulrik L Andersen, Ch Marquardt,  and Gerd Leuchs, “Generation of continuous-variable cluster states of cylindrically polarized modes,” arXiv preprint arXiv:1210.5188  (2012).
  • Ferraro et al. (2005) Alessandro Ferraro, Stefano Olivares,  and Matteo GA Paris, “Gaussian states in continuous variable quantum information,” arXiv preprint quant-ph/0503237  (2005).
  • Weedbrook et al. (2012) Christian Weedbrook, Stefano Pirandola, Raúl García-Patrón, Nicolas J. Cerf, Timothy C. Ralph, Jeffrey H. Shapiro,  and Seth Lloyd, “Gaussian quantum information,” Rev. Mod. Phys. 84, 621–669 (2012).
  • Adesso et al. (2014) Gerardo Adesso, Sammy Ragy,  and Antony R Lee, “Continuous variable quantum information: Gaussian states and beyond,” Open Systems & Information Dynamics 21, 1440001 (2014).
  • Genoni et al. (2016) Marco G. Genoni, Ludovico Lami,  and Alessio Serafini, “Conditional and unconditional gaussian quantum dynamics,” Contemporary Physics 57, 331–349 (2016).
  • (43) See Refs. Lindner and Rudolph (2009); Schwartz et al. (2016); Pichler et al. (2017) for progresses towards deterministic generation of discrete variable clusters and Refs. Spiller et al. (2006); Anders et al. (2010); Roncaglia et al. (2011); Proctor et al. (2017); Gallagher and Ferraro (2018) for alternative measurement-based approaches.
  • Lloyd and Braunstein (1999) Seth Lloyd and Samuel L. Braunstein, “Quantum computation over continuous variables,” Phys. Rev. Lett. 82, 1784–1787 (1999).
  • Menicucci et al. (2006) Nicolas C. Menicucci, Peter van Loock, Mile Gu, Christian Weedbrook, Timothy C. Ralph,  and Michael A. Nielsen, “Universal quantum computation with continuous-variable cluster states,” Phys. Rev. Lett. 97, 110501 (2006).
  • Gu et al. (2009) Mile Gu, Christian Weedbrook, Nicolas C. Menicucci, Timothy C. Ralph,  and Peter van Loock, “Quantum computing with continuous-variable clusters,” Phys. Rev. A 79, 062318 (2009).
  • Menicucci (2014) Nicolas C. Menicucci, ‘‘Fault-tolerant measurement-based quantum computing with continuous-variable cluster states,” Phys. Rev. Lett. 112, 120504 (2014).
  • (48) Notice that much effort has been devoted to counteract this issue, leading to proposals in which the necessary non-Gaussian elements can be obtained on-demand Marek et al. (2011); Yukawa et al. (2013); Marshall et al. (2015); Miyata et al. (2016); Marek et al. (2017); Takeda and Furusawa (2017); however these still require the use of quantum memories which are hard to realize Lvovsky et al. (2009); Yoshikawa et al. (2013) and that we avoid here.
  • Milburn and Woolley (2011) GJ Milburn and MJ Woolley, “An introduction to quantum optomechanics,” acta physica slovaca 61, 483–601 (2011).
  • Aspelmeyer et al. (2014) Markus Aspelmeyer, Tobias J. Kippenberg,  and Florian Marquardt, “Cavity optomechanics,” Rev. Mod. Phys. 86, 1391–1452 (2014).
  • Teufel et al. (2011) JD Teufel, Tobias Donner, Dale Li, JW Harlow, MS Allman, Katarina Cicak, AJ Sirois, Jed D Whittaker, KW Lehnert,  and Raymond W Simmonds, ‘‘Sideband cooling of micromechanical motion to the quantum ground state,” Nature 475, 359–363 (2011).
  • O’Connell et al. (2010) Aaron D O’Connell, Max Hofheinz, Markus Ansmann, Radoslaw C Bialczak, Mike Lenander, Erik Lucero, Matthew Neeley, Daniel Sank, H Wang, M Weides, et al., “Quantum ground state and single-phonon control of a mechanical resonator,” Nature 464, 697–703 (2010).
  • Noguchi et al. (2016a) Atsushi Noguchi, Rekishu Yamazaki, Manabu Ataka, Hiroyuki Fujita, Yutaka Tabuchi, Toyofumi Ishikawa, Koji Usami,  and Yasunobu Nakamura, “Ground state cooling of a quantum electromechanical system with a silicon nitride membrane in a 3d loop-gap cavity,” New Journal of Physics 18, 103036 (2016a).
  • Lei et al. (2016) C. U. Lei, A. J. Weinstein, J. Suh, E. E. Wollman, A. Kronwald, F. Marquardt, A. A. Clerk,  and K. C. Schwab, “Quantum nondemolition measurement of a quantum squeezed state beyond the 3 db limit,” Phys. Rev. Lett. 117, 100801 (2016).
  • Wollman et al. (2015) E. E. Wollman, C. U. Lei, A. J. Weinstein, J. Suh, A. Kronwald, F. Marquardt, A. A. Clerk,  and K. C. Schwab, “Quantum squeezing of motion in a mechanical resonator,” Science 349, 952–955 (2015).
  • Pirkkalainen et al. (2015) J.-M. Pirkkalainen, E. Damskägg, M. Brandt, F. Massel,  and M. A. Sillanpää, “Squeezing of quantum noise of motion in a micromechanical resonator,” Phys. Rev. Lett. 115, 243601 (2015).
  • Ockeloen-Korppi et al. (2018) CF Ockeloen-Korppi, E Damskägg, J-M Pirkkalainen, M Asjad, AA Clerk, F Massel, MJ Woolley,  and MA Sillanpää, “Stabilized entanglement of massive mechanical oscillators,” Nature 556, 478 (2018).
  • Riedinger et al. (2018) Ralf Riedinger, Andreas Wallucks, Igor Marinković, Clemens Löschnauer, Markus Aspelmeyer, Sungkun Hong,  and Simon Gröblacher, “Remote quantum entanglement between two micromechanical oscillators,” Nature 556, 473 (2018).
  • Marinkovic et al. (2018) Igor Marinkovic, Andreas Wallucks, Ralf Riedinger, Sungkun Hong, Markus Aspelmeyer,  and Simon Gröblacher, “An optomechanical bell test,” arXiv preprint arXiv:1806.10615  (2018).
  • Massel et al. (2012) Francesco Massel, Sung Un Cho, Juha-Matti Pirkkalainen, Pertti J Hakonen, Tero T Heikkilä,  and Mika A Sillanpää, “Multimode circuit optomechanics near the quantum limit,” Nature communications 3, 987 (2012).
  • Damskägg et al. (2016) Erno Damskägg, Juha-Matti Pirkkalainen,  and Mika A Sillanpää, “Dynamically creating tripartite resonance and dark modes in a multimode optomechanical system,” Journal of Optics 18, 104003 (2016).
  • Grass et al. (2016) David Grass, Julian Fesel, Sebastian G. Hofer, Nikolai Kiesel,  and Markus Aspelmeyer, “Optical trapping and control of nanoparticles inside evacuated hollow core photonic crystal fibers,” Applied Physics Letters 108, 221103 (2016)https://doi.org/10.1063/1.4953025 .
  • Nielsen et al. (2017) William Hvidtfelt Padkær Nielsen, Yeghishe Tsaturyan, Christoffer Bo Møller, Eugene S. Polzik,  and Albert Schliesser, “Multimode optomechanical system in the quantum regime,” Proceedings of the National Academy of Sciences 114, 62–66 (2017)http://www.pnas.org/content/114/1/62.full.pdf .
  • Noguchi et al. (2016b) Atsushi Noguchi, Rekishu Yamazaki, Manabu Ataka, Hiroyuki Fujita, Yutaka Tabuchi, Toyofumi Ishikawa, Koji Usami,  and Yasunobu Nakamura, “Strong coupling in multimode quantum electromechanics,” arXiv preprint arXiv:1602.01554  (2016b).
  • Bhattacharya et al. (2008) M. Bhattacharya, H. Uys,  and P. Meystre, “Optomechanical trapping and cooling of partially reflective mirrors,” Phys. Rev. A 77, 033819 (2008).
  • Thompson et al. (2008) J. D. Thompson, B. M. Zwickl, A. M. Jayich, Florian Marquardt, S. M. Girvin,  and J. G. E. Harris, ‘‘Strong dispersive coupling of a high-finesse cavity to a micromechanical membrane,” Nature 452, 72 (2008).
  • Woolley et al. (2008) M. J. Woolley, A. C. Doherty, G. J. Milburn,  and K. C. Schwab, “Nanomechanical squeezing with detection via a microwave cavity,” Phys. Rev. A 78, 062303 (2008).
  • Hertzberg et al. (2010) JB Hertzberg, T Rocheleau, T Ndukum, M Savva, AA Clerk,  and KC Schwab, “Back-action-evading measurements of nanomechanical motion,” Nature Physics 6, 213–217 (2010).
  • Rocheleau et al. (2010) T. Rocheleau, T. Ndukum, C. MacKlin, J.B. Hertzberg, A.A. Clerk,  and K.C. Schwab, “Preparation and detection of a mechanical resonator near the ground state of motion,” Nature 463, 72–75 (2010).
  • Nunnenkamp et al. (2010) A. Nunnenkamp, K. Børkje, J. G. E. Harris,  and S. M. Girvin, “Cooling and squeezing via quadratic optomechanical coupling,” Phys. Rev. A 82, 021806 (2010).
  • Purdy et al. (2010) T. P. Purdy, D. W. C. Brooks, T. Botter, N. Brahms, Z.-Y. Ma,  and D. M. Stamper-Kurn, “Tunable cavity optomechanics with ultracold atoms,” Phys. Rev. Lett. 105, 133602 (2010).
  • Sankey et al. (2010) Jack C Sankey, Cheng Yang, Benjamin M Zwickl, Andrew M Jayich,  and Jack GE Harris, “Strong and tunable nonlinear optomechanical coupling in a low-loss system,” Nature Physics 6, 707–712 (2010).
  • Hill et al. (2011) Jeffrey T. Hill, Qiang Lin, Jessie Rosenberg,  and Oskar Painter, “Mechanical trapping in a quadratically coupled optomechanical double disk,” in CLEO:2011 - Laser Applications to Photonic Applications (Optical Society of America, 2011) p. CThJ3.
  • Karuza et al. (2013) M Karuza, M Galassi, C Biancofiore, C Molinelli, R Natali, P Tombesi, G Di Giuseppe,  and D Vitali, “Tunable linear and quadratic optomechanical coupling for a tilted membrane within an optical cavity: theory and experiment,” Journal of Optics 15, 025704 (2013).
  • Flowers-Jacobs et al. (2012) NE Flowers-Jacobs, SW Hoch, JC Sankey, A Kashkanova, AM Jayich, C Deutsch, J Reichel,  and JGE Harris, “Fiber-cavity-based optomechanical device,” Applied Physics Letters 101, 221109 (2012).
  • Li et al. (2012) Hao-Kun Li, Yong-Chun Liu, Xu Yi, Chang-Ling Zou, Xue-Xin Ren,  and Yun-Feng Xiao, ‘‘Proposal for a near-field optomechanical system with enhanced linear and quadratic coupling,” Phys. Rev. A 85, 053832 (2012).
  • Hill (2013) Jeffrey Thomas Hill, Nonlinear optics and wavelength translation via cavity-optomechanics, Ph.D. thesis, California Institute of Technology (2013).
  • Doolin et al. (2014) C. Doolin, B. D. Hauer, P. H. Kim, A. J. R. MacDonald, H. Ramp,  and J. P. Davis, “Nonlinear optomechanics in the stationary regime,” Phys. Rev. A 89, 053838 (2014).
  • Kaviani et al. (2015) Hamidreza Kaviani, Chris Healey, Marcelo Wu, Roohollah Ghobadi, Aaron Hryciw,  and Paul E. Barclay, “Nonlinear optomechanical paddle nanocavities,” Optica 2, 271–274 (2015).
  • Paraïso et al. (2015) Taofiq K. Paraïso, Mahmoud Kalaee, Leyun Zang, Hannes Pfeifer, Florian Marquardt,  and Oskar Painter, “Position-squared coupling in a tunable photonic crystal optomechanical cavity,” Phys. Rev. X 5, 041024 (2015).
  • Lee et al. (2015) D Lee, M Underwood, D Mason, AB Shkarin, SW Hoch,  and JGE Harris, “Multimode optomechanical dynamics in a cavity with avoided crossings,” Nature communications 6, 6232 (2015).
  • Kim et al. (2015) Eun-jong Kim, J. R. Johansson,  and Franco Nori, “Circuit analog of quadratic optomechanics,” Phys. Rev. A 91, 033835 (2015).
  • Abdi and Hartmann (2015) Mehdi Abdi and Michael J Hartmann, “Entangling the motion of two optically trapped objects via time-modulated driving fields,” New J. Phys. 17, 013056 (2015).
  • Kalaee et al. (2016) M. Kalaee, T. K. Paraïso, H. Pfeifer,  and O. Painter, “Design of a quasi-2d photonic crystal optomechanical cavity with tunable, large x2-coupling,” Opt. Express 24, 21308–21328 (2016).
  • Fonseca et al. (2016) P. Z. G. Fonseca, E. B. Aranas, J. Millen, T. S. Monteiro,  and P. F. Barker, “Nonlinear dynamics and strong cavity cooling of levitated nanoparticles,” Phys. Rev. Lett. 117, 173602 (2016).
  • Brawley et al. (2016) GA Brawley, MR Vanner, Peter Emil Larsen, Silvan Schmid, Anja Boisen,  and WP Bowen, “Nonlinear optomechanical measurement of mechanical motion,” Nature communications 7, 10988 (2016).
  • Leijssen et al. (2017) Rick Leijssen, Giada R La Gala, Lars Freisem, Juha T Muhonen,  and Ewold Verhagen, “Nonlinear cavity optomechanics with nanomechanical thermal fluctuations,” Nature communications 8, ncomms16024 (2017).
  • Zhang et al. (2017) Lin Zhang, Fengzhou Ji, Xu Zhang,  and Weiping Zhang, “Photon–phonon parametric oscillation induced by quadratic coupling in an optomechanical resonator,” Journal of Physics B: Atomic, Molecular and Optical Physics 50, 145501 (2017).
  • Dellantonio et al. (2018) Luca Dellantonio, Oleksandr Kyriienko, Florian Marquardt,  and Anders S Sørensen, “Quantum nondemolition measurement of mechanical motion quanta,” arXiv preprint arXiv:1801.02438  (2018).
  • Kronwald et al. (2013) Andreas Kronwald, Florian Marquardt,  and Aashish A. Clerk, “Arbitrarily large steady-state bosonic squeezing via dissipation,” Phys. Rev. A 88, 063833 (2013).
  • Ikeda and Yamamoto (2013) Yusuke Ikeda and Naoki Yamamoto, “Deterministic generation of gaussian pure states in a quasilocal dissipative system,” Phys. Rev. A 87, 033802 (2013).
  • Houhou et al. (2015) Oussama Houhou, Habib Aissaoui,  and Alessandro Ferraro, “Generation of cluster states in optomechanical quantum systems,” Phys. Rev. A 92, 063843 (2015).
  • Clerk et al. (2008) A A Clerk, F Marquardt,  and K Jacobs, “Back-action evasion and squeezing of a mechanical resonator using a cavity detector,” New Journal of Physics 10, 095010 (2008).
  • Moore et al. (2017) Darren W. Moore, Oussama Houhou,  and Alessandro Ferraro, “Arbitrary multimode gaussian operations on mechanical cluster states,” Phys. Rev. A 96, 022305 (2017).
  • Zhang and Braunstein (2006) Jing Zhang and Samuel L. Braunstein, “Continuous-variable gaussian analog of cluster states,” Phys. Rev. A 73, 032318 (2006).
  • Ferrini et al. (2013) G Ferrini, J P Gazeau, T Coudreau, C Fabre,  and N Treps, “Compact gaussian quantum computation by multi-pixel homodyne detection,” New Journal of Physics 15, 093015 (2013).
  • Ghose and Sanders (2007) Shohini Ghose and Barry C. Sanders, “Non-gaussian ancilla states for continuous variable quantum computation via gaussian maps,” Journal of Modern Optics 54, 855–869 (2007).
  • Marek et al. (2011) Petr Marek, Radim Filip,  and Akira Furusawa, “Deterministic implementation of weak quantum cubic nonlinearity,” Phys. Rev. A 84, 053802 (2011).
  • Yukawa et al. (2013) Mitsuyoshi Yukawa, Kazunori Miyata, Hidehiro Yonezawa, Petr Marek, Radim Filip,  and Akira Furusawa, “Emulating quantum cubic nonlinearity,” Phys. Rev. A 88, 053816 (2013).
  • Marshall et al. (2015) Kevin Marshall, Raphael Pooser, George Siopsis,  and Christian Weedbrook, “Repeat-until-success cubic phase gate for universal continuous-variable quantum computation,” Phys. Rev. A 91, 032321 (2015).
  • Miyata et al. (2016) Kazunori Miyata, Hisashi Ogawa, Petr Marek, Radim Filip, Hidehiro Yonezawa, Jun-ichi Yoshikawa,  and Akira Furusawa, “Implementation of a quantum cubic gate by an adaptive non-gaussian measurement,” Phys. Rev. A 93, 022301 (2016).
  • Marek et al. (2017) Petr Marek, Radim Filip, Hisashi Ogawa, Atsushi Sakaguchi, Shuntaro Takeda, Jun-ichi Yoshikawa,  and Akira Furusawa, “General implementation of arbitrary nonlinear quadrature phase gates,” arXiv preprint arXiv:1708.02822  (2017).
  • Takeda and Furusawa (2017) Shuntaro Takeda and Akira Furusawa, “Universal quantum computing with measurement-induced continuous-variable gate sequence in a loop-based architecture,” Phys. Rev. Lett. 119, 120504 (2017).
  • Gottesman et al. (2001) Daniel Gottesman, Alexei Kitaev,  and John Preskill, “Encoding a qubit in an oscillator,” Phys. Rev. A 64, 012310 (2001).
  • (105) See Supplemental Material at [URL will be inserted by publisher] for more details.
  • (106) Note that the parameters , , can be tuned by varying both the bare optomechanical couplings and the driving amplitudes, while is set by changing only.
  • Carmichael (1993) Howard Carmichael, An Open System Approach to Quantum Optics, 1st ed. (Springer, 1993).
  • Gardiner and Zoller (2000) C. W. Gardiner and P. Zoller, Quantum Noise, 2nd ed. (Springer, Berlin, 2000).
  • (109) We describe the level of squeezing as  dB Gu et al. (2009).
  • Takagi and Zhuang (2018) Ryuji Takagi and Quntao Zhuang, “Convex resource theory of non-gaussianity,” arXiv preprint arXiv:1804.04669  (2018).
  • Albarelli et al. (2018) Francesco Albarelli, Marco G Genoni, Matteo GA Paris,  and Alessandro Ferraro, “Resource theory of quantum non-gaussianity and wigner negativity,” arXiv preprint arXiv:1804.05763  (2018).
  • (112) A seminal proposal to realize, via optomechanical-like interactions, states potentially useful for MBQC is given in Ref. Pirandola et al. (2006); however, there a non-linearized and probabilistic approach is considered. We also notice that the recent deterministic proposal in Ref. Brunelli et al. (2018) use similar techniques to the ones presented here but the states obtained there are not useful for MBQC (see also Brunelli and Houhou (2018) for further analysis).
  • (113) Similar tailoring techniques have been considered in the context of both continuous Menicucci (2014) and discrete variables Bermejo-Vega et al. (2017). In the latter, Gaussian and non-Gaussian gates are substituted with Clifford and non-Clifford ones.
  • (114) We should mention that our switching scheme introduced here is not only a generalisation of a previous protocol Houhou et al. (2015) for the generation of Gaussian cluster states, but also conforms with the canonical preparation of Gaussian cluster states if we restrict our selves to first sideband drivings only.
  • Marquardt et al. (2007) Florian Marquardt, Joe P. Chen, A. A. Clerk,  and S. M. Girvin, “Quantum theory of cavity-assisted sideband cooling of mechanical motion,” Phys. Rev. Lett. 99, 093902 (2007).
  • Wilson-Rae et al. (2007) I. Wilson-Rae, N. Nooshi, W. Zwerger,  and T. J. Kippenberg, “Theory of ground state cooling of a mechanical oscillator using dynamical backaction,” Phys. Rev. Lett. 99, 093901 (2007).
  • Chang et al. (2010) D. E. Chang, C. A. Regal, S. B. Papp, D. J. Wilson, J. Ye, O. Painter, H. J. Kimble,  and P. Zoller, “Cavity opto-mechanics using an optically levitated nanosphere,” Proceedings of the National Academy of Sciences 107, 1005–1010 (2010)http://www.pnas.org/content/107/3/1005.full.pdf .
  • Pino et al. (2018) H Pino, J Prat-Camps, K Sinha, B Prasanna Venkatesh,  and O Romero-Isart, “On-chip quantum interference of a superconducting microsphere,” Quantum Science and Technology 3, 025001 (2018).
  • Chuang et al. (1997) Isaac L. Chuang, Debbie W. Leung,  and Yoshihisa Yamamoto, “Bosonic quantum codes for amplitude damping,” Phys. Rev. A 56, 1114–1125 (1997).
  • Lund et al. (2008) A. P. Lund, T. C. Ralph,  and H. L. Haselgrove, “Fault-tolerant linear optical quantum computing with small-amplitude coherent states,” Phys. Rev. Lett. 100, 030503 (2008).
  • Michael et al. (2016) Marios H. Michael, Matti Silveri, R. T. Brierley, Victor V. Albert, Juha Salmilehto, Liang Jiang,  and S. M. Girvin, “New class of quantum error-correcting codes for a bosonic mode,” Phys. Rev. X 6, 031006 (2016).
  • Fukui et al. (2017) Kosuke Fukui, Akihisa Tomita,  and Atsushi Okamoto, “Analog quantum error correction with encoding a qubit into an oscillator,” Phys. Rev. Lett. 119, 180507 (2017).
  • Ofek et al. (2016) Nissim Ofek, Andrei Petrenko, Reinier Heeres, Philip Reinhold, Zaki Leghtas, Brian Vlastakis, Yehan Liu, Luigi Frunzio, SM Girvin, L Jiang, et al., “Extending the lifetime of a quantum bit with error correction in superconducting circuits,” Nature 536, 441 (2016).
  • Rosenblum et al. (2018) Serge Rosenblum, YY Gao, Philip Reinhold, Chen Wang, CJ Axline, Luigi Frunzio, SM Girvin, Liang Jiang, Mazyar Mirrahimi, MH Devoret, et al., “A cnot gate between multiphoton qubits encoded in two cavities,” Nature communications 9, 652 (2018).
  • Flühmann et al. (2018) Christa Flühmann, Thanh Long Nguyen, Matteo Marinelli, Vlad Negnevitsky, Karan Mehta,  and Jonathan Home, ‘‘Encoding a qubit in a trapped-ion mechanical oscillator,” arXiv preprint arXiv:1807.01033  (2018).
  • Lindner and Rudolph (2009) Netanel H. Lindner and Terry Rudolph, “Proposal for pulsed on-demand sources of photonic cluster state strings,” Phys. Rev. Lett. 103, 113602 (2009).
  • Schwartz et al. (2016) Ido Schwartz, Dan Cogan, Emma R Schmidgall, Yaroslav Don, Liron Gantz, Oded Kenneth, Netanel H Lindner,  and David Gershoni, “Deterministic generation of a cluster state of entangled photons,” Science 354, 434–437 (2016).
  • Pichler et al. (2017) Hannes Pichler, Soonwon Choi, Peter Zoller,  and Mikhail D. Lukin, “Universal photonic quantum computation via time-delayed feedback,” Proceedings of the National Academy of Sciences  (2017), 10.1073/pnas.1711003114http://www.pnas.org/content/early/2017/10/09/1711003114.full.pdf .
  • Spiller et al. (2006) T P Spiller, Kae Nemoto, Samuel L Braunstein, W J Munro, P van Loock,  and G J Milburn, “Quantum computation by communication,” New Journal of Physics 8, 30 (2006).
  • Anders et al. (2010) Janet Anders, Daniel K. L. Oi, Elham Kashefi, Dan E. Browne,  and Erika Andersson, “Ancilla-driven universal quantum computation,” Phys. Rev. A 82, 020301 (2010).
  • Roncaglia et al. (2011) Augusto J. Roncaglia, Leandro Aolita, Alessandro Ferraro,  and Antonio Acín, “Sequential measurement-based quantum computing with memories,” Phys. Rev. A 83, 062332 (2011).
  • Proctor et al. (2017) Timothy Proctor, Melissa Giulian, Natalia Korolkova, Erika Andersson,  and Viv Kendon, “Ancilla-driven quantum computation for qudits and continuous variables,” Phys. Rev. A 95, 052317 (2017).
  • Gallagher and Ferraro (2018) C. B. Gallagher and A. Ferraro, “Relative resilience to noise of standard and sequential approaches to measurement-based quantum computation,” Phys. Rev. A 97, 052305 (2018).
  • Lvovsky et al. (2009) Alexander I Lvovsky, Barry C Sanders,  and Wolfgang Tittel, “Optical quantum memory,” Nature photonics 3, 706 (2009).
  • Yoshikawa et al. (2013) Jun-ichi Yoshikawa, Kenzo Makino, Shintaro Kurata, Peter van Loock,  and Akira Furusawa, “Creation, storage, and on-demand release of optical quantum states with a negative wigner function,” Phys. Rev. X 3, 041028 (2013).
  • Pirandola et al. (2006) S. Pirandola, S. Mancini, D. Vitali,  and P. Tombesi, “Continuous variable encoding by ponderomotive interaction,” The European Physical Journal D - Atomic, Molecular, Optical and Plasma Physics 37, 283–290 (2006).
  • Brunelli et al. (2018) Matteo Brunelli, Oussama Houhou, Darren W Moore, Andreas Nunnenkamp, Mauro Paternostro,  and Alessandro Ferraro, “Unconditional preparation of nonclassical states via linear-and-quadratic optomechanics,” arXiv preprint arXiv:1804.00014  (2018).
  • Brunelli and Houhou (2018) Matteo Brunelli and Oussama Houhou, ‘‘Linear-and-quadratic reservoir engineering of non-Gaussian states,”  (2018)arXiv:1809.05266 .

Supplementary Material:

Unconditional measurement-based quantum computation

with optomechanical continuous variables

Supplemental Material

Oussama Houhou

Darren W. Moore

Sougato Bose

Alessandro Ferraro

Appendix A Hamiltonian derivation

Our optomechanical system consists of mechanical oscillators interacting with a cavity mode via radiation pressure, and the cavity is pumped by a multi-tone field. The full Hamiltonian of the system is given by Aspelmeyer et al. (2014)


where and are, respectively, the annihilation operator and frequency of the cavity field, , and are the frequency and (dimensionless) position and momentum quadratures of the mechanical oscillator, and is the (classical) pump of the cavity.

The cavity’s frequency becomes dependent on the mechanical positions Aspelmeyer et al. (2014). We expand in powers of up to the second order:


with and are the the position and position-squared couplings of the mechanical oscillator with the cavity field. The Hamiltonian becomes


Allowing our system to be in contact with vacuum reservoir for the cavity and thermal bath for the mechanical oscillator, leads to the following Heisenberg - Langevin equations Gardiner and Zoller (2000) for the system operators:


where and are the damping rates for the cavity mode and the mechanical oscillator, and and are the input noise operators for the cavity and mechanical oscillator respectively, satisfying the correlation relations:


with denoting the mean phonon number.

We aim to derive an effective Hamiltonian for the system involving quantum fluctuations around the (classical) fields steady states. Replacing the system operators, in equations (S4)-(S6), by their mean-fields: , and , the classical equations of motion become:


The multi-tone driving field can be written as :


where is the frequency of the pump field, and is its complex amplitude. We consider the following ansatz for the intra-cavity field at the steady state Milburn and Woolley (2011):


where the constants are the complex amplitudes of the cavity at the steady state. By substituting expression (S15) in Eq. (S12) we find:


If we assume weak coupling such that for we have


the time dependent terms in Eq. (S16) can be neglected. And if we denote by and the values of position and momentum at the steady state, it is easy to find the following:


where is the detuning of the drive with respect to the cavity.

Having obtained the steady state for all fields, we can derive a Hamiltonian of the system in terms of the quantum fluctuations around the classical steady values. First, we split the system operators into classical part and quantum fluctuation,


then we substitute (S21) in equations(S4)-(S6). Assuming a strong drive, , we find:


Equations (S22)–(S24) correspond to the following effective Hamiltonian:


where we defined and .

The explicit time-dependence of Hamiltonian (S25) can be removed by, first, going to a frame rotating with the free terms of the system where the Hamiltonian transforms to


Then, we consider a resonant drive with the cavity, with amplitude and null detuning, and four driving fields per each mechanical oscillator, with detunings and amplitudes (). By invoking the rotating wave approximation (RWA), we may write the Hamiltonian as


where we have defined and .

Appendix B Validity of the rotating wave approximation

The validity of the RWA introduced in Sec. A will be justified here. Hamiltonian (S27) is obtained from Eq. (S26) by discarding all time-dependent (counter-rotating) terms and keeping only the resonant ones. The counter-rotating terms may be written as


with the following expressions:


Now we can state the necessary conditions to safely neglect the counter-rotating terms. For the RWA to be valid, the following constraints must be met:


We study the validity of the RWA in more details for the interesting case of the preparation of the cubic phase state of a mechanical oscillator. The system consists of a cavity and one mechanical oscillator (). The full Hamiltonian of the system is again with