Accurate and efficient preparation of quantum state is a core issue in building a quantum computer. In this paper, we investigate how to prepare a certain single- or two-qubit target state from arbitrary initial states in semiconductor double quantum dots with the aid of deep reinforcement learning. Our method is based on the training of the network over numerous preparing tasks. The results show that once the network is well trained, it works for any initial states in the continuous Hilbert space. Thus repeated training for new preparation tasks is avoided. Our scheme outperforms the traditional optimization approaches based on gradient with both the higher designing efficiency and the preparation quality in discrete control space. Moreover, we find that the control trajectories designed by our scheme are robust against static and dynamic fluctuations, such as charge and nuclear noises.
We examine a single-pulse preparation of the uniform superpositional wave function, which includes all basis states, in a spin quantum computer. The effective energy spectrum and the errors generated by this pulse are studied in detail. We show that, in spite of the finite width of the energy spectrum bands, amplitude and phase errors can be made reasonably small.
We show that dissipative quantum state preparation processes can be protected against qubit dephasing by interlacing the state preparation control with dynamical decoupling (DD) control consisting of a sequence of short $π$-pulses. The inhomogeneous broadening can be suppressed to second order of the pulse interval, and the protection efficiency is nearly independent of the pulse sequence but determined by the average interval between pulses. The DD protection is numerically tested and found to be efficient against inhomogeneous dephasing on two exemplary dissipative state preparation schemes …
We report a novel Bell state preparation experiment. High-purity Bell states are prepared by using femtosecond pulse pumped \emph{nondegenerate} collinear spontaneous parametric down-conversion. The use of femtosecond pump pulse {\em does not} result in reduction of quantum interference visibility in our scheme in which post-selection of amplitudes and other traditional mechanisms, such as, using thin nonlinear crystals or narrow-band spectral filters are not used. Another distinct feature of this scheme is that the pump, the signal, and the idler wavelengths are all distinguishable, which is …
We present theoretical and experimental study of preparing maximally entangled two-photon polarization states, or Bell states, using femtosecond pulse pumped spontaneous parametric down-conversion (SPDC). First, we show how the inherent distinguishability in femtosecond pulse pumped type-II SPDC can be removed by using an interferometric technique without spectral and amplitude post-selection. We then analyze the recently introduced Bell state preparation scheme using type-I SPDC. Theoretically, both methods offer the same results, however, type-I SPDC provides experimentally superior methods …
We introduce a protocol capable of generating a general measurement operator for a mechanical resonator. The technique requires a qubit-resonator interaction and uses a coherent pulse to drive qubit transitions. This is followed by projective measurement of the qubit's energy, constraining the resonator in a state that depends on the pulse shape. The freedom to choose a pulse shape for the coherent drive enables an arbitrary position-basis measurement operator. Using this measurement operator, we outline a two pulse protocol that probabilistically generates a pure mechanical state with a desir…
Quantum projection synthesis can be used for phase-probability-distribution measurement, optical-state truncation and preparation. The method relies on interfering optical lights, which is a major challenge in experiments performed by pulsed light sources. In the pulsed regime, the time frequency overlap of the interfering lights plays a crucial role on the efficiency of the method when they have different mode structures. In this paper, the pulsed mode projection synthesis is developed, the mode structure of interfering lights are characterized and the effect of this overlap (or mode match) o…
We propose a method to deterministically prepare a desired quantum state in a one-dimensional (1D) continuum by a shaped photon pulse. This method is based on time-reverse of the quantum emission process. We show that the desired quantum state such as Dicke or timed-Dicke state can be successfully prepared with very high fidelity even if the dissipation to the environment is nonnegligible and the pulse shaping is not perfect. We also show that large quantum entanglement between emitters can be created by just a single photon pulse. This method is experimentally feasible in 1D waveguide-QED or …
A new method of preparing the pseudo-pure state of a spin system for quantum computation in liquid nuclear magnetic resonance (NMR) was put forward and demonstrated experimentally. Applying appropriately connected line-selective pulses simultaneously and a field gradient pulse techniques we acquired straightforwardly all pseudo-pure states for two qubits in a single experiment much efficiently. The signal intensity with the pseudo-pure state prepared in this way is the same as that of temporal averaging. Our method is suitable for the system with arbitrary numbers of qubits. As an example of a…
Qubits driven by resonant strong pulses are studied and a parameter regime is explored in which the dynamics can be solved in closed form. Instantaneous coherent destruction of tunneling can be seen for longer pulses, whereas shorter pulses allow a fast preparation of the qubit state. Results are compared with recent experiments of pulsed nitrogen-vacancy center spin qubits in diamond.
Quantum control has been of increasing interest in recent years, e.g. for tasks like state initialization and stabilization. Feedback-based strategies are particularly powerful, but also hard to find, due to the exponentially increased search space. Deep reinforcement learning holds great promise in this regard. It may provide new answers to difficult questions, such as whether nonlinear measurements can compensate for linear, constrained control. Here we show that reinforcement learning can successfully discover such feedback strategies, without prior knowledge. We illustrate this for state p…
We consider the problem of pulsed biexciton preparation in a quantum dot and show that a pulse-sequence with a simple on-off-on modulation can achieve complete preparation of the target state faster than the commonly used constant and hyperbolic secant pulses. The durations of the pulses composing the sequence are obtained from the solution of a transcendental equation. Furthermore, using numerical optimal control, we demonstrate that for a wide range of values of the maximum pulse amplitude, the proposed pulse-sequence prepares the biexciton state in the numerically obtained minimum time, for…
A novel method for preparation of a single photon in temporally-delocalized entangled modes is proposed and analyzed. We show that two single-photon pulses propagating in a driven nonabsorbing medium with different group velocities are temporally split under parametric interaction into well-separated pulses. In consequence, the single-photon "time-bin-entangled" states are generated with a programmable entanglement easily controlled by driving field intensity. The experimental study of nonclassical features and nonlocality of generated states by means of balanced homodyne tomography is discuss…
Reinforcement learning has been widely used in many problems, including quantum control of qubits. However, such problems can, at the same time, be solved by traditional, non-machine-learning methods, such as stochastic gradient descent and Krotov algorithms, and it remains unclear which one is most suitable when the control has specific constraints. In this work, we perform a comparative study on the efficacy of three reinforcement learning algorithms: tabular Q-learning, deep Q-learning, and policy gradient, as well as two non-machine-learning methods: stochastic gradient descent and Krotov …
We propose a method based on deep reinforcement learning that efficiently prepares a quantum many-body pure state in thermal or prethermal equilibrium. The main physical intuition underlying the method is that the information on the equilibrium states can be efficiently encoded/extracted by focusing on only a few local observables, relying on the typicality of equilibrium states. Instead of resorting to the expensive preparation protocol that adopts global features such as the quantum state fidelity, we show that the equilibrium states can be efficiently prepared only by learning the expectati…
Quantum computing has seen tremendous progress in the past years. Due to the implementation complexity and cost, the future path of quantum computation is strongly believed to delegate computational tasks to powerful quantum servers on cloud. Universal blind quantum computing (UBQC) provides the protocol for the secure delegation of arbitrary quantum computations, and it has received significant attention. However, a great challenge in UBQC is how to transmit quantum state over long distance securely and reliably. Here, we solve this challenge by proposing and demonstrating a resource-efficien…
Within the framework of exact quantum electrodynamics in dispersing and absorbing media, we have studied the quantum state of the radiation emitted from an initially in the upper state prepared two-level atom in a high-$Q$ cavity, including the regime where the emitted photon belongs to a wave packet that simultaneously covers the areas inside and outside the cavity. For both continuing atom--field interaction and short-term atom--field interaction, we have determined the spatio-temporal shape of the excited outgoing wave packet and calculated the efficiency of the wave packet to carry a one-p…
Pulsed optomechanical measurements enable squeezing, non-classical state creation and backaction-free sensing. We demonstrate pulsed measurement of a cryogenic nanomechanical resonator with record precision close to the quantum regime. We use these to prepare thermally squeezed and purified conditional mechanical states, and to perform full state tomography. These demonstrations exploit large photon-phonon coupling in a nanophotonic cavity to reach a single-pulse imprecision of 9 times the mechanical zero-point amplitude $x_\mathrm{zpf}$. We study the effect of other mechanical modes which lim…
We report on a simple method to prepare optical pulses with exponentially rising envelope on the time scale of a few ns. The scheme is based on the exponential transfer function of a fast transistor, which generates an exponentially rising envelope that is transferred first on a radio frequency carrier, and then on a coherent cw laser beam with an electro-optical phase modulator (EOM). The temporally shaped sideband is then extracted with an optical resonator and can be used to efficiently excite a single Rb-87 atom.
We report on the excitation of single molecules via narrow zero-phonon transitions using short laser pulses. By monitoring the Stokes-shifted fluorescence, we studied the excited state population as a function of the delay time, laser intensity, and frequency detuning. A pi-pulse excitation was demonstrated with merely 500 photons, and 5 Rabi cycles were achieved at higher excitation powers. Our findings are in good agreement with theoretical calculations and provide a first step toward coherent manipulation of the electronic states of single molecules with few photons.
The energy states in semiconductor quantum dots are discrete as in atoms, and quantum states can be coherently controlled with resonant laser pulses. Long coherence times allow the observation of Rabi-flopping of a single dipole transition in a solid state device, for which occupancy of the upper state depends sensitively on the dipole moment and the excitation laser power. We report on the robust preparation of a quantum state using an optical technique that exploits rapid adiabatic passage from the ground to an excited state through excitation with laser pulses whose frequency is swept throu…
Preparation of quantum state lies at the heart of quantum information processing. The greedy algorithm provides a potential method to effectively prepare quantum states. However, the standard greedy algorithm, in general, cannot take the global maxima and instead becomes stuck on a local maxima. Based on the standard greedy algorithm, in this paper we propose a revised version to design dynamic pulses to realize universal quantum state preparation, i.e., preparing any arbitrary state from another arbitrary one. As applications, we implement this scheme to the universal preparation of single- a…
We consider performing adiabatic rapid passage (ARP) using frequency-swept driving pulses to excite a collection of interacting two-level systems. Such a model arises in a wide range of many-body quantum systems, such as cavity QED or quantum dots, where a nonlinear component couples to light. We analyze the one-dimensional case using the Jordan-Wigner transformation, as well as the mean field limit where the system is described by a Lipkin-Meshkov-Glick Hamiltonian. These limits provide complementary insights into the behavior of many-body systems under ARP, suggesting our results are general…
Deep reinforcement learning for near-deterministic preparation of cubic- and quartic-phase gates in photonic quantum computing
Department of Physics, University of Virginia（弗吉尼亚大学物理系）；Charles L. Brown Department of Electrical and Computer Engineering, University of Virginia（弗吉尼亚大学查尔斯·L·布朗电气与计算机工程系）
Quantum information processing often requires the preparation of arbitrary quantum states, such as all the states on the Bloch sphere for two-level systems. While numerical optimization can prepare individual target states, they lack the ability to find general solutions that work for a large class of states in more complicated quantum systems. Here, we demonstrate global quantum control by preparing a continuous set of states with deep reinforcement learning. The protocols are represented using neural networks, which automatically groups the protocols into similar types, which could be useful…
We have investigated theoretically and experimentally a method for preparing Dicke states in trapped atomic ions. We consider a linear chain of $N$ ion qubits that is prepared in a particular Fock state of motion, $|m>$. The $m$ phonons are removed by applying a laser pulse globally to the $N$ qubits, and converting the motional excitation to $m$ flipped spins. The global nature of this pulse ensures that the $m$ flipped spins are shared by all the target ions in a state that is a close approximation to the Dicke state $\D{N}{m}$. We calculate numerically the fidelity limits of the protocol an…
Loss-aware pulse sequence optimization for generating photonic Fock states
Friedrich Schiller University Jena（耶拿弗里德里希·席勒大学）；Paderborn University（帕德博恩大学）；Fraunhofer Institute for Applied Optics and Precision Engineering IOF（弗劳恩霍夫应用光学和精密工程研究所 IOF）；Institute for Photonic Quantum Systems (PhoQS)（光子量子系统研究所 (PhoQS)）
A fundamental prerequisite for the implementation of linear optical quantum computation is a source of single-photon wavepackets capable of high-visibility interference in scalable networks. These conditions can be met with micro-structured waveguides in conjunction with ultra-short classical timing pulses. By exploiting a novel type-II phasematching configuration we demonstrate a waveguided single photon source exhibiting a conditional detection efficiency exceeding 51% (which corresponds to a preparation efficiency of 85%) and extraordinarily high detection rates of up to 8.5x10^5 coincidenc…
We examine fully coherent two-pulse propagation in a lambda-type medium, under two-photon resonance conditions and including inhomogeneous broadening. We examine both the effects of short pulse preparation and the effects of medium preparation. We contrast cases in which the two pulses have matched envelopes or not, and contrast cases in which ground state coherence is present or not. We find that an extended interpretation of the Area Theorem for single-pulse self-induced transparency (SIT) is able to unify two-pulse propagation scenarios, including some aspects of electromagnetically-induced…
In the Gaussian-modulated coherent states (GMCS) quantum key distribution (QKD) protocol, Alice prepares quantum states \emph{actively}: for each transmission, Alice generates a pair of Gaussian-distributed random numbers, encodes them on a weak coherent pulse using optical amplitude and phase modulators, and then transmits the Gaussian-modulated weak coherent pulse to Bob. Here we propose a \emph{passive} state preparation scheme using a thermal source. In our scheme, Alice splits the output of a thermal source into two spatial modes using a beam splitter. She measures one mode locally using …
We consider the two-step method [A. R. Kuzmak and V. M. Tkachuk, Phys. Lett. A 378 (2014) 1469] for preparation of an arbitrary quantum gate on two spins with anisotropic Heisenberg interaction. At the first step, the system evolves during some period of time. At the second step, we apply pulsed magnetic field individually to each spin. We obtain the conditions for realization of SWAP, iSWAP, $\sqrt{SWAP}$ and entangled gates. Finally, we consider the implementation of this method on the physical system of ultracold atoms in optical lattice.
We propose a potentially practical scheme, in combination with the Bell-state analyzer [Zhang et al., Phys. Rev. A 73, 014301 (2006)], to generate Bell states for two electron spins confined, respectively, in two distant fullerenes. To this end, we consider the endohedral fullerenes staying in single walled carbon nanotubes (SWCNTs) and employ auxiliary mobile electrons and selective microwave pulses. The application and the experimental feasibility of our scheme are discussed.
Theoretical Quantum Information Processing (QIP) has matured from the use of qubits to the use of qudits (systems having states> 2). Where as most of the experimental implementations have been performed using qubits, little experimental work has been carried out using qudits as yet. In this paper we demonstrate experimental realization of a qutrit system by nuclear magnetic resonance (NMR), utilizing deuterium (spin-1) nuclei partially oriented in liquid crystalline phase. Preparation of pseudopure states and implementation of unitary operations are demonstrated in this single-qutrit system, u…
We experimentally demonstrate an ultrafast method for preparing spin states of donor-bound electrons in GaAs with single laser pulses. Each polarization state of a preparation pulse has a direct mapping onto a spin state, with bijective correspondence between the Poincaré-sphere (for photon polarization) and Bloch-sphere (for spin) state representations. The preparation is governed by a stimulated Raman process and occurs orders of magnitude faster than the spontaneous emission and spin dephasing. Similar dynamics governs our ultrafast optical Kerr detection of the spin coherence, thus getting…
The dynamics of a system that is initially correlated with an environment is almost always non-Markovian. Hence it is important to characterise such initial correlations experimentally and witness them in physically realistic settings. One such setting is weak-field phase control, where chemical reactions are sought to be controlled by the phase of shaped weak laser pulses. In this manuscript, we show how weak quantum controllability can be combined with quantum preparations to witness initial correlations between the system and the environment. Furthermore we show how weak field phase control…
We propose a variational quantum algorithm to prepare ground states of 1D lattice quantum Hamiltonians specifically tailored for programmable quantum devices where interactions among qubits are mediated by Quantum Data Buses (QDB). For trapped ions with the axial Center-Of-Mass (COM) vibrational mode as single QDB, our scheme uses resonant sideband optical pulses as resource operations, which are potentially faster than off-resonant couplings and thus less prone to decoherence. The disentangling of the QDB from the qubits by the end of the state preparation comes as a byproduct of the variatio…
We propose a scheme to generate double electromagnetically induced transparency and optimal cross-phase modulation for two slow, copropagating pulses with matched group velocities in a single species of atom, namely 87 Rb. A single pump laser is employed and a homogeneous magnetic field is utilized to avoid cancellation effects through the nonlinear Zeeman effect. We suggest a feasible preparational procedure for the atomic initial state to achieve matched group velocities for both signal fields.
We investigate both experimentally and theoretically a simple yet more robust and flexible alternative to Rabi oscillation-type biexciton preparation protocols traditionally used for semiconductor quantum dots. The quantum dot is excited by a strong laser pulse positively detuned from the two-photon resonance yielding an on demand initialization of the biexciton state by making use of the phonon-induced thermalization of the photon dressed states. It is shown that for excitation pulses in the picosecond range, a stable and high fidelity of up to $f_{XX}=0.98\pm 0.01$ is reached. Notably, the g…
