YOLSO: You Only Look Small Object

Small object detection is challenging and far from satisfactory. Most general object detectors suffer from two critical issues with small objects: (1) Feature extractor based on classification network cannot express the characteristics of small objects reasonably due to insufficient appearance information of targets and a large amount of background interference around them. (2) The detector requires a much higher location accuracy for small objects than for general objects. This paper proposes an effective and efficient small object detector YOLSO to address the above problems. For feature representation, we analyze the drawbacks in previous backbones and present a Half-Space Shortcut(HSSC) module to build a background-aware backbone. Furthermore, a coarse-to-fine Feature Pyramid Enhancement(FPE) module is introduced for layer-wise aggregation at a granular level to enhance the semantic discriminability. For loss function, we propose an exponential L1 loss to promote the convergence of regression, and a focal IOU loss to focus on prime samples with high classification confidence and high IOU. Both of them significantly improves the location accuracy of small objects. The proposed YOLSO sets state-of-the-art results on two typical small object datasets, MOCOD and VeDAI, at a speed of over 200 FPS. In the meantime, it also outperforms the baseline YOLOv3 by a wide margin on the common COCO dataset.     

用于电弧检测的电力专用SoC设计

针对传统的电弧电路故障检测结果不准确的问题,设计用于电弧检测的SoC系统,并且在55nm工艺下进行流片验证。采用包含两种结构的模数转换器的片上电压源,设计了锁相环以及复位电路,精度最高可达8.67 bit。验证结果表明,本设计可提高电弧检测的准确性。     

Tabular data: Deep learning is not all you need

A key element in solving real-life data science problems is selecting the types of models to use. Tree ensemble models (such as XGBoost) are usually recommended for classification and regression problems with tabular data. However, several deep learning models for tabular data have recently been proposed, claiming to outperform XGBoost for some use cases. This paper explores whether these deep models should be a recommended option for tabular data by rigorously comparing the new deep models to XGBoost on various datasets. In addition to systematically comparing their performance, we consider the tuning and computation they require. Our study shows that XGBoost outperforms these deep models across the datasets, including the datasets used in the papers that proposed the deep models. We also demonstrate that XGBoost requires much less tuning. On the positive side, we show that an ensemble of deep models and XGBoost performs better on these datasets than XGBoost alone.     

疫情下电磁场理论课程的网络教学

本文主要总结了新冠疫情期间作者的电磁场理论课程在线教学经验。对比分析了录播和直播的优缺点后,选择录播教学方式。基于超星网络教学平台,展示了录播网络教学的具体措施,包括网上答疑和学习效果检查以及在线批改作业等。给出了网络教学可以为线下教学继续使用的方法和手段,为疫情结束后的正常教学提供了新的网络教学补充措施。     

基于误差估值累加开环校正的诱导式欺骗检测方法

诱导式卫星欺骗干扰可诱导航空器逐渐偏离预定航迹，难以被发现，因此及时有效地检测干扰是飞行安全的保障。在现有紧组合导航体制基础上，设计了一种基于误差估值累加开环校正的紧组合导航结构，并证明了其性能与传统闭环校正紧组合导航性能等效。在此结构中，将紧组合导航系统与自适应序贯概率比检测方法结合，提出了一种基于误差估值累加开环校正的诱导式欺骗检测方法，融合紧组合导航信息与其他不受欺骗影响的导航信息，构建欺骗检测统计量进行诱导式欺骗检测。仿真结果表明，开环校正结构可避免随时间累加的惯性导航系统误差所导致的组合导航滤波器发散问题，同时欺骗检测方法可进一步提高算法对“最坏”情形下微小诱导式欺骗的检测效果。     

Machine fault diagnosis with small sample based on variational information constrained generative adversarial network

In actual engineering scenarios, limited fault data leads to insufficient model training and over-fitting, which negatively affects the diagnostic performance of intelligent diagnostic models. To solve the problem, this paper proposes a variational information constrained generative adversarial network (VICGAN) for effective machine fault diagnosis. Firstly, by incorporating the encoder into the discriminator to map the deep features, an improved generative adversarial network with stronger data synthesis capability is established. Secondly, to promote the stable training of the model and guarantee better convergence, a variational information constraint technique is utilized, which constrains the input signals and deep features of the discriminator using the information bottleneck method. In addition, a representation matching module is added to impose restrictions on the generator, avoiding the mode collapse problem and boosting the sample diversity. Two rolling bearing datasets are utilized to verify the effectiveness and stability of the presented network, which demonstrates that the presented network has an admirable ability in processing fault diagnosis with few samples, and performs better than state-of-the-art approaches.     

使能未来工厂的5G能力综述

5G系统将移动通信服务从移动电话、移动宽带和大规模机器通信扩展到新的应用领域，即所谓对通信服务有特殊要求的垂直领域。对使能未来工厂的5G能力进行了全面的分析总结，包括弹性网络架构、灵活频谱、超可靠低时延通信、时间敏感网络、安全和定位，而弹性网络架构又包括对网络切片、非公共网络、5G局域网和边缘计算的支持。希望从广度到深度，对相关的理论及技术应用做透彻、全面的梳理，对其挑战做清晰的总结，从而为相关研究和工程技术人员提供借鉴。     

Applications of sketches in network traffic measurement: A survey

Accurate and timely network traffic measurement is essential for network status monitoring, network fault analysis, network intrusion detection, and network security management. With the rapid development of the network, massive network traffic brings severe challenges to network traffic measurement. However, existing measurement methods suffer from many limitations for effectively recording and accurately analyzing big-volume traffic. Recently, sketches, a family of probabilistic data structures that employ hashing technology for summarizing traffic data, have been widely used to solve these problems. However, current literature still lacks a thorough review on sketch-based traffic measurement methods to offer a comprehensive insight on how to apply sketches for fulfilling various traffic measurement tasks. In this paper, we provide a detailed and comprehensive review on the applications of sketches in network traffic measurement. To this end, we classify the network traffic measurement tasks into four categories based on the target of traffic measurement, namely cardinality estimation, flow size estimation, change anomaly detection, and persistent spreader identification. First, we briefly introduce these four types of traffic measurement tasks and discuss the advantages of applying sketches. Then, we propose a series of requirements with regard to the applications of sketches in network traffic measurement. After that, we perform a fine-grained classification for each sketch-based measurement category according to the technologies applied on sketches. During the review, we evaluate the performance, advantages and disadvantages of current sketch-based traffic measurement methods based on the proposed requirements. Through the thorough review, we gain a number of valuable implications that can guide us to choose and design proper traffic measurement methods based on sketches. We also review a number of general sketches that are highly expected in modern network systems to simultaneously perform multiple traffic measurement tasks and discuss their performance based on the proposed requirements. Finally, through our serious review, we summarize a number of open issues and identify several promising research directions.     

Structure-sensitive semantic matching for aggregate question answering over knowledge base

Aggregate question answering essentially returns answers for given questions by obtaining query graphs with unique dependencies between values and corresponding objects. Word order dependency, as the key to uniquely identify dependency of the query graph, reflects the dependencies between the words in the question. However, due to the semantic gap caused by the expression difference between questions encoded with word vectors and query graphs represented with logical formal elements, it is not trivial to match the correct query graph for the question. Most existing approaches design more expressive query graphs for complex questions and rank them just by directly calculating their similarities, ignoring the semantic gap between them. In this paper, we propose a novel Structure-sensitive Semantic Matching(SSM) approach that learns aligned representations of dependencies in questions and query graphs to eliminate their gap. First, we propose a cross-structure matching module to bridge the gap between two modalities(i.e., textual question and query graph). Then, we propose an entropy-based gated AQG filter to remove the structural noise caused by the uncertainty of dependencies. Finally, we present a two-channel query graph representation that fuses the semantics of abstract structure and grounding content of the query graph explicitly. Experimental results show that SSM could learn aligned representations of questions and query graphs to eliminate the gaps between their dependencies, and improves up to 12% (F1 score) on aggregation questions of two benchmark datasets.