Beijing University of Posts and Telecommunications

About: Beijing University of Posts and Telecommunications is a education organization based out in Beijing, Beijing, China. It is known for research contribution in the topics: MIMO & Quality of service. The organization has 39576 authors who have published 41525 publications receiving 403759 citations. The organization is also known as: BUPT.

Heterogeneous Graph Attention Networks for Semi-supervised Short Text Classification

Abstract: Short text classification has found rich and critical applications in news and tweet tagging to help users find relevant information. Due to lack of labeled training data in many practical use cases, there is a pressing need for studying semi-supervised short text classification. Most existing studies focus on long texts and achieve unsatisfactory performance on short texts due to the sparsity and limited labeled data. In this paper, we propose a novel heterogeneous graph neural network based method for semi-supervised short text classification, leveraging full advantage of few labeled data and large unlabeled data through information propagation along the graph. In particular, we first present a flexible HIN (heterogeneous information network) framework for modeling the short texts, which can integrate any type of additional information as well as capture their relations to address the semantic sparsity. Then, we propose Heterogeneous Graph ATtention networks (HGAT) to embed the HIN for short text classification based on a dual-level attention mechanism, including node-level and type-level attentions. The attention mechanism can learn the importance of different neighboring nodes as well as the importance of different node (information) types to a current node. Extensive experimental results have demonstrated that our proposed model outperforms state-of-the-art methods across six benchmark datasets significantly.

Cooperative UAV Cluster-Assisted Terrestrial Cellular Networks for Ubiquitous Coverage

Abstract: Unmanned aerial vehicles (UAVs), featured by flexible configuration, robust deployment, and line-of-sight links, has a great potential to provide ubiquitous wireless coverage and high-speed transmission. In this paper, we aim to analyze the coverage performance of UAV-assisted terrestrial cellular networks, where partially energy-harvesting-powered caching UAVs are randomly deployed in the 3-D space with a minimum and maximum altitude, i.e., $H_{l}$ and $H_{h}$ . A novel cooperative UAV clustering scheme is proposed to offload ground mobile terminals (GMTs) from ground cellular base stations to cooperative UAV clusters. A cooperative UAV cluster is developed within a cylinder with projection centered on a GMT, based on their energy states, the cached contents, and the cell loads. With tractable Poisson point process and Gamma approximation, explicit expressions for the successful transmission probabilities are obtained. A theoretical analysis reveals that the cooperative probability of a UAV and the offloading probability of a GMT have bell-shaped relation with respect to the radius of the cylinder and the cache hit probability (the matching probability of a content request and content cache). Numerical results are provided to demonstrate the impacts of the system parameters on the cooperative UAV cluster. The results also give the optimal average altitude ( ${H_{l}+H_{h}}/{2}$ ) and altitude difference ( $H_{h}-H_{l}$ ) in maximizing the coverage performance with the proposed cooperative transmission scheme.

Geometry Guided Convolutional Neural Networks for Self-Supervised Video Representation Learning

Abstract: It is often laborious and costly to manually annotate videos for training high-quality video recognition models, so there has been some work and interest in exploring alternative, cheap, and yet often noisy and indirect training signals for learning the video representations. However, these signals are still coarse, supplying supervision at the whole video frame level, and subtle, sometimes enforcing the learning agent to solve problems that are even hard for humans. In this paper, we instead explore geometry, a grand new type of auxiliary supervision for the self-supervised learning of video representations. In particular, we extract pixel-wise geometry information as flow fields and disparity maps from synthetic imagery and real 3D movies, respectively. Although the geometry and high-level semantics are seemingly distant topics, surprisingly, we find that the convolutional neural networks pre-trained by the geometry cues can be effectively adapted to semantic video understanding tasks. In addition, we also find that a progressive training strategy can foster a better neural network for the video recognition task than blindly pooling the distinct sources of geometry cues together. Extensive results on video dynamic scene recognition and action recognition tasks show that our geometry guided networks significantly outperform the competing methods that are trained with other types of labeling-free supervision signals.

On the non-planar dust-ion-acoustic waves in cosmic dusty plasmas with transverse perturbations

Abstract: Dust-ion-acoustic waves in a cosmic dusty plasma is investigated, in spherical geometry with both azimuthal and zenith perturbations existing. (3+1)-dimensional spherical Kadomtsev-Petviashvili ((3+1)DsKP) model is constructed with symbolic computation, 4th-ordered and with variable coefficients. Auto-Backlund transformation and (3+1)DsKP nebulons are analytically obtained for such a generic model. Astromechanical and physical implications are discussed, of the supernova-shell-type expanding bright and Saturn-F-ring-type expanding dark (3+1)DsKP nebulons. Possibly observable nebulonic effects are proposed for future cosmic experiments.

A self-powered deep-ultraviolet photodetector based on an epitaxial Ga2O3/Ga:ZnO heterojunction

Abstract: A deep-ultraviolet photodetector (DUV PD) which can function independently of an external power supply is urgently desired for the next-generation photodetection applications from the viewpoint of being diminutive, convenient, and power saving. In this work, by introducing a lattice compatible semiconductor Ga:ZnO, a high quality β-Ga2O3/Ga:ZnO heterojunction based DUV PD is achieved using laser molecular beam epitaxy. The obtained device could operate in a self-powered mode with an excellent wavelength selectivity, a high ON/OFF ratio, a high DUV/visible rejection ratio, and a high stability under 254 nm light illumination. The physical mechanism responsible for the observation is discussed based on the photogenerated electron–hole pairs separated in the depletion region under the built-in electric field. Our work may provide a new insight into further high performance self-sufficient DUV PD applications.