Attention-relation network for mobile phone screen defect classification via a few samples

How to use a few defect samples to complete the defect classification is a key challenge in the production of mobile phone screens.An attention-relation network for the mobile phone screen defect classification is proposed in this paper.The architecture of the attention-relation network contains two modules:a feature extract module and a feature metric module.Different from other few-shot models,an attention mechanism is applied to metric learning in our model to measure the distance between features,so as to pay attention to the correlation between features and suppress unwanted information.Besides,we combine dilated convolution and skip connection to extract more feature information for follow-up processing.We validate attention-relation network on the mobile phone screen defect dataset.The experimental results show that the classification accuracy of the attentionrelation network is 0.9486 under the 5-way 1-shot training strategy and 0.9039 under the 5-way 5-shot setting.It achieves the excellent effect of classification for mobile phone screen defects and outperforms with dominant advantages.

BARN:Behavior-Aware Relation Network for multi-label behavior detection in socially housed macaques

Quantification of behaviors in macaques provides crucial support for various scientific disciplines,including pharmacology,neuroscience,and ethology.Despite recent advancements in the analysis of macaque behavior,research on multi-label behavior detection in socially housed macaques,including consideration of interactions among them,remains scarce.Given the lack of relevant approaches and datasets,we developed the Behavior-Aware Relation Network(BARN)for multi-label behavior detection of socially housed macaques.Our approach models the relationship of behavioral similarity between macaques,guided by a behavior-aware module and novel behavior classifier,which is suitable for multi-label classification.We also constructed a behavior dataset of rhesus macaques using ordinary RGB cameras mounted outside their cages.The dataset included 65?913 labels for19 behaviors and 60?367 proposals,including identities and locations of the macaques.Experimental results showed that BARN significantly improved the baseline SlowFast network and outperformed existing relation networks.In conclusion,we successfully achieved multilabel behavior detection of socially housed macaques with both economic efficiency and high accuracy.

New International Partner Network Launched to Further Sino-US Business Relations

"The network will foster newrelationship between US andChinese small and medium-size companies in 14 key busi-ness centers, generating newopportunities for US SMEs inthe China market and prosper-ity for both our great nations,"said Tim Hauser.

Optimal paths planning in dynamic transportation networks with random link travel times

A theoretical study was conducted on finding optimal paths in transportation networks where link travel times were stochastic and time-dependent(STD). The methodology of relative robust optimization was applied as measures for comparing time-varying, random path travel times for a priori optimization. In accordance with the situation in real world, a stochastic consistent condition was provided for the STD networks and under this condition, a mathematical proof was given that the STD robust optimal path problem can be simplified into a minimum problem in specific time-dependent networks. A label setting algorithm was designed and tested to find travelers' robust optimal path in a sampled STD network with computation complexity of O(n2+n·m). The validity of the robust approach and the designed algorithm were confirmed in the computational tests. Compared with conventional probability approach, the proposed approach is simple and efficient, and also has a good application prospect in navigation system.

HTTP-sCAN:Detecting HTTP-Flooding Attack by Modeling Multi-Features of Web Browsing Behavior from Noisy Web-Logs

HTTP-flooding attack disables the victimized web server by sending a large number of HTTP Get requests.Recent research tends to detect HTTP-flooding with the anomaly-based approaches,which detect the HTTP-flooding by modeling the behavior of normal web surfers.However,most of the existing anomaly-based detection approaches usually cannot filter the web-crawling traces from unknown searching bots mixed in normal web browsing logs.These web-crawling traces can bias the base-line profile of anomaly-based schemes in their training phase,and further degrade their detection performance.This paper proposes a novel web-crawling tracestolerated method to build baseline profile,and designs a new anomaly-based HTTP-flooding detection scheme(abbr.HTTP-sCAN).The simulation results show that HTTP-sCAN is immune to the interferences of unknown webcrawling traces,and can detect all HTTPflooding attacks.

Data Mapping Principles and Algorithm for Self-Configurable Systems

With the challenges brought by the expansion of network scale,as well as the diversity of the equipments and the complexity of network protocols,many self-configurable systems have been proposed combining formal specification and model finding techniques.In this paper,we pay more attention to formal specifications of network information,i.e.,exploring principles and algorithm to map network information(topology,devices and status,etc.) to Alloy specifications.We first model network information in relational form,which is easy to realize because of the structured feature of network information in nature.Then we map the relational data to Alloy specifications according to our novel data mapping principles and algorithm.Based on the transition of relational data,it is possible to automatically map network information to Alloy specifications.We evaluate our data mapping principles and algorithm by applying them to a practical application scenario.The results illustrate that we can find a model for the task within a tolerant time interval,which implies that our novel approach can convert relational data to Alloy specifications correctly and efficiently.

Geomorphic indices and relative tectonic uplift in the Guerrero sector of the Mexican forearc

Tectonically active areas,such as forearc regions,commonly show contrasting relief,differential tectonic uplift,variations in erosion rates,in river incision,and in channel gradient produced by ongoing tectonic deformation.Thus,information on the tectonic activity of a defined area could be derived via landscape analysis.This study uses topography and geomorphic indices to extract signals of ongoing tectonic deformation along the Mexican subduction forearc within the Guerrero sector.For this purpose,we use field data,topographical data,knickpoints,the ratio of volume to area（Rva）.the stream-length gradient index（St）,and the normalized channel steepness index（k_（sn））.The results of the applied landscape analysis reveal considerable variations in relief,topography and geomorphic indices values along the Guerrero sector of the Mexican subduction zone.We argue that the reported differences are indicative of tectonic deformation and of variations in relative tectonic uplift along the studied forearc.A significant drop from central and eastern parts of the study area towards the west in values of R_（VA）（from ~500 to^300）,St（from ~500 to ca.400）,maximum St（from ~1500-2500 to ~ 1000） and k_（sn）（from ~150 to ~100） denotes a decrease in relative tectonic uplift in the same direction.We suggest that applied geomorphic indices values and forearc topography are independent of climate and lithology.Actual mechanisms responsible for the observed variations and inferred changes in relative forearc tectonic uplift call for further studies that explain the physical processes that control the forearc along strike uplift variations and that determine the rates of uplift.The proposed methodology and results obtained through this study could prove useful to scientists who study the geomorphology of forearc regions and active subduction zones.

Transformer fault diagnosis based on relational teacher-student network

The analysis of dissolved gas in oil can provide an important basis for transformer fault diagnosis.In order to improve the accuracy of transformer fault diagnosis,a method based on the relational teacher-student network(R-TSN)is proposed by analyzing the relationship between the dissolved gas in the oil and the fault type.R-TSN replaces the original hard labels with soft labels,and uses it to measure the similarity between different samples in the space,to a certain extent,it can obtain the hidden feature information in the samples,and clarify the classification boundary.Through the identification experiment,the effect of R-TSN diagnosis model is analyzed,and the influence of the compound fault of discharge and thermal on the diagnosis model is studied.This paper compares R-TSN with support vector machines(SVMs),decision trees and multilayer perceptron models in transformer fault diagnosis.Experimental results show that R-TSN has better performance than the above methods.After adding compound faults in the sample set,the accuracy rate can still reach 86.0%.

In Silico Investigation of Agonist Activity of a Structurally Diverse Set of Drugs to hPXR Using HM-BSM and HM-PNN

The human pregnane X receptor（hPXR） plays a critical role in the metabolism, transport and clearance of xenobiotics in the liver and intestine. The hPXR can be activated by a structurally diverse of drugs to initiate clinically relevant drug-drug interactions. In this article, in silico investigation was performed on a structurally diverse set of drugs to identify critical structural features greatly related to their agonist activity towards h PXR. Heuristic method（HM）-Best Subset Modeling（BSM） and HM-Polynomial Neural Networks（PNN） were utilized to develop the linear and non-linear quantitative structure-activity relationship models. The applicability domain（AD） of the models was assessed by Williams plot. Statistically reliable models with good predictive power and explain were achieved（for HM-BSM, r^2=0.881, q^2_（LOO）=0.797, q^2_（EXT）=0.674; for HM-PNN, r^2=0.882, q^2_（LOO）=0.856, q^2_（EXT）=0.655）. The developed models indicated that molecular aromatic and electric property, molecular weight and complexity may govern agonist activity of a structurally diverse set of drugs to h PXR.

Graph-Segmenter:graph transformer with boundary-aware attention for semantic segmentation

Thetransformer-based semantic segmentation approaches,which divide the image into different regions by sliding windows and model the relation inside each window,have achieved outstanding success.However,since the relation modeling between windows was not the primary emphasis of previous work,it was not fully utilized.To address this issue,we propose a Graph-Segmenter,including a graph transformer and a boundary-aware attention module,which is an effective network for simultaneously modeling the more profound relation between windows in a global view and various pixels inside each window as a local one,and for substantial low-cost boundary adjustment.Specifically,we treat every window and pixel inside the window as nodes to construct graphs for both views and devise the graph transformer.The introduced boundary-awareattentionmoduleoptimizes theedge information of the target objects by modeling the relationship between the pixel on the object's edge.Extensive experiments on three widely used semantic segmentation datasets(Cityscapes,ADE-20k and PASCAL Context)demonstrate that our proposed network,a Graph Transformer with Boundary-aware Attention,can achieve state-of-the-art segmentation performance.

Understanding Social Relationships with Person-Pair Relations

Social relationship understanding infers existing social relationships among individuals in a given scenario,which has been demonstrated to have a wide range of practical value in reality.However,existing methods infer the social relationship of each person pair in isolation,without considering the context-aware information for person pairs in the same scenario.The context-aware information for person pairs exists extensively in reality,that is,the social relationships of different person pairs in a simple scenario are always related to each other.For instance,if most of the person pairs in a simple scenario have the same social relationship,“friends”,then the other pairs have a high probability of being“friends”or other similar coarse-level relationships,such as“intimate”.This context-aware information should thus be considered in social relationship understanding.Therefore,this paper proposes a novel end-to-end trainable Person-Pair Relation Network(PPRN),which is a GRU-based graph inference network,to first extract the visual and position information as the person-pair feature information,then enable it to transfer on a fully-connected social graph,and finally utilizes different aggregators to collect different kinds of person-pair information.Unlike existing methods,the method—with its message passing mechanism in the graph model—can infer the social relationship of each person-pair in a joint way(i.e.,not in isolation).Extensive experiments on People In Social Context(PISC)-and People In Photo Album(PIPA)-relation datasets show the superiority of our method compared to other methods.

Convergence analysis of distributed Kalman filtering for relative sensing networks

We study the distributed Kalman filtering problem in relative sensing networks with rigorous analysis.The relative sensing network is modeled by an undirected graph while nodes in this network are running homogeneous dynamical models. The sufficient and necessary condition for the observability of the whole system is given with detailed proof. By local information and measurement communication, we design a novel distributed suboptimal estimator based on the Kalman filtering technique for comparison with a centralized optimal estimator. We present sufficient conditions for its convergence with respect to the topology of the network and the numerical solutions of n linear matrix inequality（LMI） equations combining system parameters. Finally, we perform several numerical simulations to verify the effectiveness of the given algorithms.