Quantum causality violates classical intuitions of cause and effect and is a unique quantum feature different from other quantum phenomena such as entanglement and quantum nonlocality. In order to avoid the detection loophole in quantum causality, we
initiate the study of the detection efficiency requirement for observing quantum causality. We first show that previous classical causal inequalities require detection efficiency at least 95.97% (89.44%) to show violation with quantum (nonsignaling) correlations. Next we derive a classical causal inequality I_{222} and show that it requires lower detection efficiency to be violated, 92.39% for quantum correlations and 81.65% for nonsignaling correlations, hence substantially reducing the requirement on detection. Then we extend this causal inequality to the case of multiple measurement settings and analyze the corresponding detection efficiency. After that, we show that previous quantum causal inequalities require detection efficiency at least 94.29% to violate with nonsignaling correlations. We subsequently derive a quantum causal bound J_{222} that has a lower detection efficiency requirement of 91.02% for violation with nonsignaling correlations. Our work paves the way towards an experimental demonstration of quantum causality and shows that causal inequalities significantly differ from Bell inequalities in terms of the detection efficiency requirement.
Low-dimensional excitonic materials have inspired much interest owing to their novel physical and technological prospects. In particular, those with strong in-plane anisotropy are among the most intriguing but short of general analyses. We establish
the universal functional form of the anisotropic dispersion in the small $k$ limit for 2D dipolar excitonic systems. While the energy is linearly dispersed in the direction parallel to the dipole in-plane, the perpendicular direction is dispersionless up to linear order, which can be explained by the quantum interference effect of the interaction among the constituents of 1D subsystems. The anisotropic dispersion results in a $E^{sim0.5}$ scaling of the system density of states and predicts unique spectroscopic signatures including: (1) disorder-induced absorption linewidth, $W(sigma)simsigma^{2.8}$, with $sigma$ the disorder strength, (2) temperature dependent absorption linewidth, $W(T)sim T^{s+1.5}$, with $s$ the exponent of the environment spectral density, and (3) the out-of-plane angular $theta$ dependence of the peak splittings in absorption spectra, $Delta E(theta)proptosin^2theta$. These predictions are confirmed quantitatively with numerical simulations of molecular thin films and tubules.
In the past few years, convolutional neural networks (CNNs) have achieved milestones in medical image analysis. Especially, the deep neural networks based on U-shaped architecture and skip-connections have been widely applied in a variety of medical
image tasks. However, although CNN has achieved excellent performance, it cannot learn global and long-range semantic information interaction well due to the locality of the convolution operation. In this paper, we propose Swin-Unet, which is an Unet-like pure Transformer for medical image segmentation. The tokenized image patches are fed into the Transformer-based U-shaped Encoder-Decoder architecture with skip-connections for local-global semantic feature learning. Specifically, we use hierarchical Swin Transformer with shifted windows as the encoder to extract context features. And a symmetric Swin Transformer-based decoder with patch expanding layer is designed to perform the up-sampling operation to restore the spatial resolution of the feature maps. Under the direct down-sampling and up-sampling of the inputs and outputs by 4x, experiments on multi-organ and cardiac segmentation tasks demonstrate that the pure Transformer-based U-shaped Encoder-Decoder network outperforms those methods with full-convolution or the combination of transformer and convolution. The codes and trained models will be publicly available at https://github.com/HuCaoFighting/Swin-Unet.
The method of deep learning has achieved excellent results in improving the performance of robotic grasping detection. However, the deep learning methods used in general object detection are not suitable for robotic grasping detection. Current modern
object detectors are difficult to strike a balance between high accuracy and fast inference speed. In this paper, we present an efficient and robust fully convolutional neural network model to perform robotic grasping pose estimation from an n-channel input image of the real grasping scene. The proposed network is a lightweight generative architecture for grasping detection in one stage. Specifically, a grasping representation based on Gaussian kernel is introduced to encode training samples, which embodies the principle of maximum central point grasping confidence. Meanwhile, to extract multi-scale information and enhance the feature discriminability, a receptive field block (RFB) is assembled to the bottleneck of our grasping detection architecture. Besides, pixel attention and channel attention are combined to automatically learn to focus on fusing context information of varying shapes and sizes by suppressing the noise feature and highlighting the grasping object feature. Extensive experiments on two public grasping datasets, Cornell and Jacquard demonstrate the state-of-the-art performance of our method in balancing accuracy and inference speed. The network is an order of magnitude smaller than other excellent algorithms while achieving better performance with an accuracy of 98.9$%$ and 95.6$%$ on the Cornell and Jacquard datasets, respectively.
Spatial symmetries of quantum systems leads to important effects in spectroscopy, such as selection rules and dark states. Motivated by the increasing strength of light-matter interaction achieved in recent experiments, we investigate a set of dynami
cally-generalized symmetries for quantum systems, which are subject to a strong periodic driving. Based on Floquet response theory, we study rotational, particle-hole, chiral and time-reversal symmetries and their signatures in spectroscopy, including symmetry-protected dark states (spDS), a Floquet band selection rule (FBSR), and symmetry-induced transparency (siT). Specifically, a dynamical rotational symmetry establishes dark state conditions, as well as selection rules for inelastic light scattering processes; a particle-hole symmetry introduces dark states for symmetry related Floquet states and also a transparency effect at quasienergy crossings; chiral symmetry and time-reversal symmetry alone do not imply dark state conditions, but can be combined to the particle-hole symmetry. Our predictions reveal new physical phenomena when a quantum system reaches the strong light-matter coupling regime, important for superconducting qubits, atoms and molecules in optical or plasmonic field cavities, and optomechanical systems.
The question of how quantum coherence facilitates energy transfer has been intensively debated in the scientific community. Since natural and artificial light-harvesting units operate under the stationary condition, we address this question via a non
-equilibrium steady-state analysis of a molecular dimer irradiated by incoherent sunlight and then generalize the key predictions to arbitrarily-complex exciton networks. The central result of the steady-state analysis is the coherence-flux-efficiency relation:$eta=csum_{i eq j}F_{ij}kappa_j=2csum_{i eq j}J_{ij}{rm Im}[{rho}_{ij}]kappa_j$ with $c$ the normalization constant. In this relation, the first equality indicates that energy transfer efficiency $eta$ is uniquely determined by the trapping flux, which is the product of flux $F$ and branching ratio $kappa$ for trapping at the reaction centers, and the second equality indicates that the energy transfer flux $F$ is equivalent to quantum coherence measured by the imaginary part of the off-diagonal density matrix, i.e., $F_{ij}=2J_{ij}{rm Im}[{rho}_{ij}]$. Consequently, maximal steady-state coherence gives rise to optimal efficiency. The coherence-flux-efficiency relation holds rigorously and generally for any exciton networks of arbitrary connectivity under the stationary condition and is not limited to incoherent radiation or incoherent pumping. For light-harvesting systems under incoherent light, non-equilibrium energy transfer flux (i.e. steady-state coherence) is driven by the breakdown of detailed balance and by the quantum interference of light-excitations and leads to the optimization of energy transfer efficiency. It should be noted that the steady-state coherence or, equivalently, efficiency is the combined result of light-induced transient coherence, inhomogeneous depletion, and system-bath correlation, and is thus not necessarily correlated with quantum beatings.
Robotic grasping plays an important role in the field of robotics. The current state-of-the-art robotic grasping detection systems are usually built on the conventional vision, such as RGB-D camera. Compared to traditional frame-based computer vision
, neuromorphic vision is a small and young community of research. Currently, there are limited event-based datasets due to the troublesome annotation of the asynchronous event stream. Annotating large scale vision dataset often takes lots of computation resources, especially the troublesome data for video-level annotation. In this work, we consider the problem of detecting robotic grasps in a moving camera view of a scene containing objects. To obtain more agile robotic perception, a neuromorphic vision sensor (DAVIS) attaching to the robot gripper is introduced to explore the potential usage in grasping detection. We construct a robotic grasping dataset named Event-Stream Dataset with 91 objects. A spatio-temporal mixed particle filter (SMP Filter) is proposed to track the led-based grasp rectangles which enables video-level annotation of a single grasp rectangle per object. As leds blink at high frequency, the Event-Stream dataset is annotated in a high frequency of 1 kHz. Based on the Event-Stream dataset, we develop a deep neural network for grasping detection which consider the angle learning problem as classification instead of regression. The method performs high detection accuracy on our Event-Stream dataset with 93% precision at object-wise level. This work provides a large-scale and well-annotated dataset, and promotes the neuromorphic vision applications in agile robot.
We study the effect of an applied magnetic field on the nonequilibrium transport properties of a general cubic quantum network described by a tight-binding Hamiltonian with specially designed couplings to the leads that preserve open-system symmetrie
s. We demonstrate that the symmetry of open systems can be manipulated by the direction of the magnetic field. Starting with all the symmetries preserved in absence of a field, the anisotropic and isotropic fields systematically break the symmetries, influencing all nonequilibrium properties. For simple cubic systems, we are able to identify the steady states that comprise of pure states, bath-dependent states (nonequilibrium steady states), and also nonphysical states. As an application, we show numerically for large cubic networks that the symmetry breaking can control nonequilibrium currents and that different environmental interactions can lead to novel features which can be engineered in artificial super-lattices and cold atoms.
The Aharanov-Bohm (AB) effect, which predicts that a magnetic field strongly influences the wave function of an electrically charged particle, is investigated in a three site system in terms of the quantum control by an additional dephasing source. T
he AB effect leads to a non-monotonic dependence of the steady-state current on the gauge phase associated with the molecular ring. This dependence is sensitive to site energy, temperature, and dephasing, and can be explained using the concept of the dark state. Although the phase effect vanishes in the steady-state current for strong dephasing, the phase dependence remains visible in an associated waiting-time distribution, especially at short times. Interestingly, the phase rigidity (i.e., the symmetry of the AB phase) observed in the steady-state current is now broken in the waiting-time statistics, which can be explained by the interference between transfer pathways.
Classical correlation can be locked via quantum means--quantum data locking. With a short secret key, one can lock an exponentially large amount of information, in order to make it inaccessible to unauthorized users without the key. Quantum data lock
ing presents a resource-efficient alternative to one-time pad encryption which requires a key no shorter than the message. We report experimental demonstrations of quantum data locking scheme originally proposed by DiVincenzo et al. [Phys. Rev. Lett. 92, 067902 (2004)] and a loss-tolerant scheme developed by Fawzi, Hayde, and Sen [J. ACM. 60, 44 (2013)]. We observe that the unlocked amount of information is larger than the key size in both experiments, exhibiting strong violation of the incremental proportionality property of classical information theory. As an application example, we show the successful transmission of a photo over a lossy channel with quantum data (un)locking and error correction.
