IQC Student Summer Conference

A key prediction of quantum field theory that has yet to be tested experimentally is the existence of correlations between different regions in a quantum field. It is hypothesized that this phenomenon can be measured using the entanglement harvesting protocol, a process by which entanglement between detectors is induced due to their interaction with a quantum field in its vacuum state. Entanglement harvesting has been extensively researched using perturbative methods. However, experimental proposals for realizing this protocol using superconducting qubits utilize setups beyond the limits of perturbation theory. Furthermore, non-perturbative studies are very limited to particular scenarios often unfit for modeling the regimes of current experiments. Here we present results on entanglement harvesting using non-perturbative methods. We investigate the breakdown of perturbation theory as well as the non-perturbative behaviour of harvesting in realistic experimental regimes

In this presentation, I show a memory efficient numerical method designed for use in high dimensional, time dependent partial differential equations. In particular, I will show how these methods are adapted for application to the correlation functions of a scalar quantum field, and how they can be used to numerically determine the solution to biscalar functions containing two independent time coordinates.

It is widely anticipated that a quantized theory of gravity will admit quantum spacetime configurations that are described by a superposition of semiclassical spacetimes. However, in the absence of such a complete theory of quantum gravity, can we learn anything about how such states might behave? Recent developments led by Foo et al., propose an operational approach to this problem by describing the response of a first-quantized two-level quantum detector coupled to a quantum-controlled superposition of spacetimes.  Using this operational approach, we investigate what happens to an accelerated detector in such a superposition of spacetimes. We find that previously observed resonance peaks in the response function (occurring at rational values of the quantized spacetime parameter) are accentuated by the acceleration. Moreover, we provide the first explicit analysis of detector thermalization in superposed spacetimes. If time permits, I will comment on how this extension of the Unruh effect relates to previous work that found non-thermal responses for detectors travelling along superpositions of accelerated trajectories in a fixed spacetime.

Superpositions of spacetimes have received considerable attention from the Relativistic Quantum Information community. The standard RQI protocol involves calculating the response of an Unruh-DeWitt detector that is coupled to a quantum field whose background spacetime exists in a quantum superposition of geometries. In this work, we propose a method to simulate such a superposition of spacetimes using optical media. The approach builds on the established analogy between the electromagnetic field in an arbitrary spacetime and the electromagnetic field in an equivalent optical medium. Since permittivity determines the speed of light propagation in an optical medium, controlling it allows us to mimic different spacetime geometries. By analyzing an optical cavity under quantum control—where its permittivity depends on the state of a quantum system—we analyze in-principle measurable quantities in a scenario describing a superposition of effective metrics. We draw analogies between our setup and the standard RQI protocol, and outline how such an approach could be used to simulate quantum superpositions of spacetimes in a laboratory setting.

While ideal quantum key distribution (QKD) systems are well-understood, practical implementations face various vulnerabilities, such as side-channel attacks resulting from device imperfections. To address these issues, one usually adjusts the security proof to incorporate the device imperfections or modifies the experimental implementation to patch the loophole. However, in the process of certifying QKD systems, evaluators must be able to verify that the manufacturer's claims are valid. For example, current security proofs for decoy-state BB84 protocols either assume uniform phase randomization of Alice’s signals, which can be compromised by practical limitations and attacks like injection locking, or rely on a (partially) characterized phase distribution. In this talk, I will present an experimental method to evaluate QKD systems against injection-locking attacks using a heterodyne detection setup, exemplifying the evaluation process. The methods presented are source-agnostic and can be used to evaluate general QKD systems against injection-locking attacks.

First introduced by Jessie MacWilliams in her 1962 dissertation, weight enumerators are combinatorial objects that count the Hamming weight of codewords in classical error-correcting codes (and of stabilizers in quantum stabilizer codes). In 1998, Eric Rains used the weight enumerator framework to demonstrate that the distance of any quantum error-correcting code is at most 2⌊n/6⌋ + 2, building upon a series of works on classical weight enumerators by Sloane, Mallows, and Conway. Rains’ bound remains one of the best general distance bounds for quantum codes. Weight enumerators have found applications in quantum channel capacity, magic state distillation, and entanglement distillation. In this talk, we explore the intimate connections between weight enumerators, code distance, depolarizing channels, and quantum channel capacity.

In this talk, we report on work in progress on the problem of unifying the graph-theoretic and logical view points of contextuality and the connections of these approaches to Lambda polytopes. Much of the motivation of this work stems from the fact that although all these approaches have been used with substantial success, the setting in which they are best applied differ vastly. Our starting point is the celebrated result by Howard et al. which established a connection between contextuality and magic state distillation. We provide an alternative proof of this result entirely using logical Bell inequalities. More precisely, we give a complete description of stabilizer measurements on 2-qudits and derive logical Bell inequalities which identify the faces of the simulable polytope of a single qudit. This approach is a step towards understanding a concrete connection between the logical and graph-theoretic points of view of contextuality and offers a way to relate them to Lambda polytopes.

Coupling the small size and high-quality of nanomechanical resonators with low-loss electromagnetic cavities for efficient readout, cavity optomechanical systems have been used to realize exquisite sensors of quantities such as mass, force, and torque. However, owing to their low resonant frequencies, mechanical sensors are often limited by thermal noise. Therefore, reducing this noise and enhancing the coupling strength of a cavity optomechanical sensor is essential for improving its sensitivity. Recent efforts to cool infrared cavity optomechanical torque sensors have been limited by the thermal heating associated with these high energy measurement photons. Here we propose a method to circumvent these detrimental heating effects using a superconducting optomechanical torque sensor operating at mm-wave frequencies. To enhance the sensitivity of these devices, we adjust the geometry of the device to optimize its mechanical resonant frequency and moment of inertia. We perform simulations of various sensor geometries to assess their impact on the device’s torque sensitivity and coupling strength. From the simulation results, we identify the optimal configuration that simultaneously minimizes added noise and maximizes coupling strength. Future cavity-optomechanical torque sensors will be fabricated based off of these designs and measured in cryogenic settings, with the ultimate goal of reaching the standard quantum limit of optomechanical torque sensing.

We demonstrate a full implementation of the BB84 Quantum Key Distribution (QKD) protocol using a high performance Purcell-enhanced semiconductor quantum dot source capable of deterministic single photon emission. We dynamically selected Alice's states used in the protocol using a custom built electro-optic modulator, driven by a random sequence from a quantum random number generator. This is achieved at an 80 MHz repetition rate, and we observe a mean photon number of 0.0013 with an anti-bunching value of 0.03. Through our setup, we achieve a low QBER of 2.9%. In order to demonstrate the performance of our protocol, we show viable key rates against both IID and coherent attacks by an eavesdropper using both state of the art numerical and analytic security proof techniques, while being able to consider device imperfections on both the source and detector sides. The viability of these results demonstrates the potential of deterministic photon sources for QKD, and bridges high-performance photonic hardware with rigorous security analyses.

Quantum systems, particularly those undergoing non-unitary dynamics, are difficult to simulate classically, and require careful consideration of the most appropriate representations of the involved quantum states and operators. When simulating noisy quantum circuits, or general quantum channels, considered approaches, including direct density matrix simulations, tensor-networks, and truncated operator expansions, have advantages and disadvantages at capturing various classical and quantum correlations that arise due to evolution times and noise scales. Here, we propose a novel simulation technique, where general quantum mixed states are represented in terms of their associated distributions of local, informationally-complete positive-valued-measurement outcomes (POVM), and such distributions are represented efficiently using Matrix Product States (POVM-MPS). Given the POVM-MPS now represent probability densities of measurements and not probability amplitudes of states, upon application of quantum channels, its bond dimensions are truncated using a combination of canonical singular value and novel non-negative matrix factorization methods, allowing efficient approximate evolution of such systems. To demonstrate these methods, we simulate several noisy quantum circuits, and characterize various fidelity and entanglement measures as a function of circuit depth and noise scale. Such studies highlight advantages and disadvantages of the proposed methods, and demonstrate that there exists underlying, persistent difficulties of non-unitary simulation that any numerical method will encounter.

We present a method of voltage pulse design for the optimal control of spin qubits in a linear array of quantum dots. Voltage pulses are reverse-engineered from the voltage-dependent spin Hamiltonian parameters: g-factor deviations δgi(V, W) and exchange couplings Ji(V, W), when their pulse shapes S(t) are constrained to ensure time-ordered evolution. We show that a single numerical integration of a system of ODEs of type d[V, W]/dS = F(V, W) enables one to reconstruct voltage pulses Vi(S(t)), Wi(S(t)) for any shape function chosen for the spin Hamiltonian controls. The procedure yields pulses for single-qubit rotations in the global ESR field, SWAPk/2, or Control-Phase gates, with theoretically perfect unitary fidelities. We also present a strategy to systematically reduce the number of necessary voltage controls such as using a frequency-modulated rotating frame. These approaches open a pathway to simplifying experimental devices without compromising their controllability. Remarkably, the complexity of our method scales polynomially with the number of physical controls, which enables one to utilize it efficiently for the universal unitary control of large qubit arrays.