We use the exceptional point in Hopfield-Bogoliubov matrix to find the phase transition points in the bosonic system. In many previous jobs, the excitation energy vanished at the critical point. It can be stated equivalently that quantum critical poi
nt is obtained when the determinant of Hopfield-Bogoliubov matrix vanishes. We analytically obtain the Hopfield-Bogoliubov matrix corresponding to the general quadratic Hamiltonian. For single-mode system the appearance of the exceptional point in Hopfield-Bogoliubov matrix is equivalent to the disappearance of the determinant of Hopfield-Bogoliubov matrix. However, in multi-mode bosonic system, they are not equivalent except in some special cases. For example, in the case of perfect symmetry, that is, swapping any two subsystems and keeping the total Hamiltonian invariable, the exceptional point and the degenerate point coincide all the time when the phase transition occurs. When the exceptional point and the degenerate point do not coincide, we find a significant result. With the increase of two-photon driving intensity, the normal phase changes to the superradiant phase, then the superradiant phase changes to the normal phase, and finally the normal phase changes to the superradiant phase.
Dissipative quantum Rabi System, a finite-component system composed of a single two-level atom interacting with an optical cavity field mode, exhibits a quantum phase transition, which can be exploited to greatly enhance the estimation precision of u
nitary parameters (frequency and coupling strength). Here, using the quantum Langevin equation, standard mean field theory and adiabatic elimination, we investigate the quantum thermometry of a thermal bath surrounding the atom with quantum optical probes. With the increase of coupling strength between the atom and the cavity field, two kinds of singularities can be observed. One type of singularity is the exceptional point (EP) in the anti-parity-time (anti-$mathcal{PT}$) symmetrical cavity field. The other type of singularity is the critical point (CP) of phase transition from the normal to superradiant phase. We show that the optimal measurement precision occurs at the CP, instead of the EP. And the direct photon detection represents an excellent proxy for the optimal measurement near the CP. In the case where the thermal bath to be tested is independent of the extra thermal bath interacting with the cavity field, the estimation precision of the temperature always increases with the coupling strength. Oppositely, if the thermal bath to be tested is in equilibrium with the extra bath interacting with the cavity field, noises that suppress the information of the temperature will be introduced when increasing the coupling strength unless it is close to the CP.
Long sequence time-series forecasting (LSTF) has become increasingly popular for its wide range of applications. Though superior models have been proposed to enhance the prediction effectiveness and efficiency, it is reckless to neglect or underestim
ate one of the most natural and basic temporal properties of time-series. In this paper, we introduce a new baseline for LSTF, the historical inertia (HI), which refers to the most recent historical data-points in the input time series. We experimentally evaluate the power of historical inertia on four public real-word datasets. The results demonstrate that up to 82% relative improvement over state-of-the-art works can be achieved even by adopting HI directly as output.
This paper investigates the stochastic distributed nonconvex optimization problem of minimizing a global cost function formed by the summation of $n$ local cost functions. We solve such a problem by involving zeroth-order (ZO) information exchange. I
n this paper, we propose a ZO distributed primal-dual coordinate method (ZODIAC) to solve the stochastic optimization problem. Agents approximate their own local stochastic ZO oracle along with coordinates with an adaptive smoothing parameter. We show that the proposed algorithm achieves the convergence rate of $mathcal{O}(sqrt{p}/sqrt{T})$ for general nonconvex cost functions. We demonstrate the efficiency of proposed algorithms through a numerical example in comparison with the existing state-of-the-art centralized and distributed ZO algorithms.
The security of object detection systems has attracted increasing attention, especially when facing adversarial patch attacks. Since patch attacks change the pixels in a restricted area on objects, they are easy to implement in the physical world, es
pecially for attacking human detection systems. The existing defenses against patch attacks are mostly applied for image classification problems and have difficulty resisting human detection attacks. Towards this critical issue, we propose an efficient and effective plug-in defense component on the YOLO detection system, which we name Ad-YOLO. The main idea is to add a patch class on the YOLO architecture, which has a negligible inference increment. Thus, Ad-YOLO is expected to directly detect both the objects of interest and adversarial patches. To the best of our knowledge, our approach is the first defense strategy against human detection attacks. We investigate Ad-YOLOs performance on the YOLOv2 baseline. To improve the ability of Ad-YOLO to detect variety patches, we first use an adversarial training process to develop a patch dataset based on the Inria dataset, which we name Inria-Patch. Then, we train Ad-YOLO by a combination of Pascal VOC, Inria, and Inria-Patch datasets. With a slight drop of $0.70%$ mAP on VOC 2007 test set, Ad-YOLO achieves $80.31%$ AP of persons, which highly outperforms $33.93%$ AP for YOLOv2 when facing white-box patch attacks. Furthermore, compared with YOLOv2, the results facing a physical-world attack are also included to demonstrate Ad-YOLOs excellent generalization ability.
We consider contact manifolds equipped with Carnot-Caratheodory metrics, and show that the Rumin complex is respected by Sobolev mappings: Pansu pullback induces a chain mapping between the smooth Rumin complex and the distributional Rumin complex. A
s a consequence, the Rumin flat complex -- the analog of the Whitney flat complex in the setting of contact manifolds -- is bilipschitz invariant. We also show that for Sobolev mappings between general Carnot groups, Pansu pullback induces a chain mapping when restricted to a certain differential ideal of the de Rham complex. Both results are applications of the Pullback Theorem from our previous paper.
We investigate the advantage of coherent superposition of two different coded channels in quantum metrology. In a continuous variable system, we show that the Heisenberg limit $1/N$ can be beaten by the coherent superposition without the help of inde
finite causal order. And in parameter estimation, we demonstrate that the strategy with the coherent superposition can perform better than the strategy with quantum textsc{switch} which can generate indefinite causal order. We analytically obtain the general form of estimation precision in terms of the quantum Fisher information and further prove that the nonlinear Hamiltonian can improve the estimation precision and make the measurement uncertainty scale as $1/N^m$ for $mgeq2$. Our results can help to construct a high-precision measurement equipment, which can be applied to the detection of coupling strength and the test of time dilation and the modification of the canonical commutation relation.
We investigate the parameter estimation in a magnon-cavity-magnon coupled system. PT symmetrical two magnons system can be formed in the gain magnetic materials by the adiabatic elimination of the cavity field mode. We show that the optimal estimatio
n will not appear at the exceptional point due to that the quantum fluctuations are the strongest at the exceptional point. Moreover, we demonstrate that the measurements at the exceptional point tend to be optimal with the increase of prepared time. And the direct photon detection is the optimal measurement for the initial state in the vacuum input state. For the open PT symmetrical two magnons system, the quantum fluctuations will greatly reduce the degree of entanglement. Finally, we show that a higher estimated magnetic sensitivity can be obtained by measuring the frequency of one magnon in the PT symmetrical two magnons system.
We show that in an $m$-step Carnot group, a probability measure with finite $m^{th}$ moment has a well-defined Buser-Karcher center-of-mass, which is a polynomial in the moments of the measure, with respect to exponential coordinates. Using this, we
improve the main technical result of our previous paper concerning Sobolev mappings between Carnot groups; as a consequence, a number of rigidity and structural results from that paper hold under weaker assumptions on the Sobolev exponent. We also give applications to quasiregular mappings, extending earlier work in the $2$-step case to general Carnot groups.
Every day, more people are becoming infected and dying from exposure to COVID-19. Some countries in Europe like Spain, France, the UK and Italy have suffered particularly badly from the virus. Others such as Germany appear to have coped extremely wel
l. Both health professionals and the general public are keen to receive up-to-date information on the effects of the virus, as well as treatments that have proven to be effective. In cases where language is a barrier to access of pertinent information, machine translation (MT) may help people assimilate information published in different languages. Our MT systems trained on COVID-19 data are freely available for anyone to use to help translate information published in German, French, Italian, Spanish into English, as well as the reverse direction.
