Search

Recently, many efforts have been devoted to promoting the Emotion-Cause Pair Extraction (ECPE) task, as jointly extracting emotions and their causes is considered more helpful than only identifying the emotions in many applications. Among the existing efforts, end-to-end approaches are getting popular as the main trend, while others like pipeline models have been overlooked due to their potential issues of cascading errors. Nevertheless, the advantages of the pipeline models, such as logically dividing a complicated task into multiple easier subtasks, are underestimated and not well exploited. Moreover, the existing end-to-end approaches fail to capture the implicit co-occurrence or exclusion patterns between multiple pairs of emotions and causes since they are extracted independently. In view of these limitations, we propose a novel two-stage model to address the ECPE task and incorporate reinforcement learning (RL) to tackle the cascading error issue. In particular, our two-stage model first detects emotion clauses and then recognizes cause clauses for each detected emotion clause sequentially. By representing the error of each decision as an explicit reward, our model clearly knows how the error at each stage affects the final performance, hence the model can adjust itself for better performance. Furthermore, the sequential prediction enables our model to use the results achieved in the previous stages as auxiliary information in the subsequent stages. Extensive experiments on the benchmark dataset demonstrate the effectiveness of our proposed two-stage model, and the ablation comparison shows the promising effect of reducing cascading errors by incorporating RL. [ABSTRACT FROM AUTHOR]

We propose a weakly-coupled ultra-flattened chromatic dispersion few-mode photonic crystal fiber (FM-PCF) supporting 10 vector modes (HE11a, HE11b, TM01, HE21a, HE21b, TE01, HE31a, HE31b, EH11a, EH11b) which works over a large range of wavelength across the C+L band. We investigate the chromatic dispersion, minimum effective refractive index difference (minΔneff), confinement loss, bending loss and differential mode delay (DMD) of the proposed fiber under the influences of design parameters by using the finite element method (FEM). The circular holes in the first ring are replaced by elliptical holes to obtain flattened-chromatic dispersion. And an extra small rectangular defected hole is introduced in the fiber core order to achieve ultra-flattened chromatic dispersion and large minΔneff. The simulated results show that the designed weakly-coupled FM-PCF can obtain ultra-flattened (dispersion slope < 1.8 × 10−4 ns/km/nm2) and large minΔneff (> 1.6 × 10−4) in the wavelength range of 1520 nm to 1640 nm. And the proposed weakly-coupled FM-PCF also has ultra-low confinement loss (<10−9dB/km) and good bending resistance. In summary, the proposed weakly-coupled FM-PCF has the potential applications for large-capacity MDM communication systems. [ABSTRACT FROM AUTHOR]

In this letter, we apply code division multiple access (CDMA) technology to an integrated visible light communication and positioning (VLCP) system, which can recover the original user data and obtain the positioning information simultaneously. The two unipolar code sets with flexible construction methods, zero cross correlation (ZCC) code set and optical zero correlation zone (OZCZ) code set, are adopted and evaluated in the proposed system. Compared with the traditional spreading code set, these two code sets can directly suit to the VLCP system with higher transmission efficiency. In addition, the results show that zero correlation properties of OZCZ code set can effectively overcome non-perfect synchronization effects in the quasi-synchronous (QS) system. In a space of 6 m $\times $ 6 m $\times $ 5 m, the bit error rate (BER) of $1.58\times {10^{-3}}$ and average positioning error of 2.73 cm are successfully achieved in the synchronous system, while the BER of $1.60\times {10^{-3}}$ and average positioning error of 3.62 cm are realized in the QS system. [ABSTRACT FROM AUTHOR]

In this article, a novel vehicle-mounted magnetic suspension flywheel battery with a virtual inertia spindle is proposed, which has the advantages of high integration, superior energy storage characteristics, high safety, and stability. Different from the traditional flywheel battery with inertia spindle structure through the motor and flywheel, the novel flywheel battery has no inertia spindle, but has a pillar integrated with the flywheel instead. The pillar does not run through the flywheel, magnetic bearings and motor at all, therefore it does not play the role of supporting the above key components like the traditional inertia spindle, but just for convenient detection and safety protection. Therefore, the existence of the pillar is not the inertia spindle, but just looks like it, so it is called virtual inertia spindle. The unique flywheel structure after optimization can make the energy storage characteristics and the integration of flywheel battery improve obviously. In addition, because of the spherical element, the flywheel supported by the high-integration magnetic bearing has the advantage of not being easily affected by external interference produced by running conditions. To prove the validity of the proposed flywheel battery topology and correctness of theoretical analysis, the stiffness and performance tests are carried out. The experiments results show that the flywheel battery system exhibits superior antiinterference characteristic. [ABSTRACT FROM AUTHOR]

Data in many real-world applications such as social networks, users shopping behaviors, and inter-item relationships can be represented as graphs. Graph Neural Networks (GNNs) have shown great success in learning meaningful representations for graphs by inherently integrating node information and topological structure. Data in social recommendations can also be denotes as graph data in the form of user-user social graphs and user-item graphs. In addition, the relationships between items can be denoted as item-item graphs. GNNs provide an unprecedented opportunity to advance social recommendations. However, there are tremendous challenges in building GNNs-based social recommendations where (1) users (items) are simultaneously involved in the user-item graph and user-user social graph (item-item graph); (2) user-item graphs not only contain user-item interactions but also include users’ opinions on items; and (3) the nature of social relations are heterogeneous among users. In this paper, we propose a novel graph neural network framework (GraphRec+) for social recommendations, which is able to coherently model graph data in order to learn better user and item representations. Specifically, we introduce a principled approach for jointly capturing interactions and opinions in the user-item graph and also propose an attention mechanism to differentiate the heterogeneous strengths of social relations. Comprehensive experiments on three real-world datasets show the effectiveness of the proposed framework. [ABSTRACT FROM AUTHOR]

Inferring the bandwidth of internal links from the bandwidth of end-to-end paths, so-termed bandwidth tomography, is a long-standing open problem in the network tomography literature. The difficulty is due to the fact that no existing mathematical tool is directly applicable to solve the inverse problem with a set of $min$ -equations. We systematically tackle this challenge by designing a polynomial-time algorithm that returns the exact bandwidth value for all identifiable links and the tightest error bound for unidentifiable links for a given set of measurement paths. When the measurement paths are not given in advance, we prove the hardness of building measurement paths that can be used for deriving the global tightest error bounds for unidentifiable links. Accordingly, we develop a reinforcement learning (RL) approach for measurement path construction, that utilizes the special knowledge in bandwidth tomography and integrates both offline training and online prediction. Evaluation results with real-world ISP topology as well as simulated networks demonstrate that compared to other path construction methods, Random and Diversity Preferred, our RL-based path construction method can build measurement paths that result in a much smaller average error bound of the link bandwidth. [ABSTRACT FROM AUTHOR]