Minimax entropy-based co-training for fault diagnosis of blast furnace

Due to the problems of few fault samples and large data fluctuations in the blast furnace(BF)ironmaking process,some transfer learning-based fault diagnosis methods are proposed.The vast majority of such methods perform distribution adaptation by reducing the distance between data distributions and applying a classifier to generate pseudo-labels for self-training.However,since the training data is dominated by labeled source domain data,such classifiers tend to be weak classifiers in the target domain.In addition,the features generated after domain adaptation are likely to be at the decision boundary,resulting in a loss of classification performance.Hence,we propose a novel method called minimax entropy-based co-training(MMEC)that adversarially optimizes a transferable fault diagnosis model for the BF.The structure of MMEC includes a dual-view feature extractor,followed by two classifiers that compute the feature's cosine similarity to representative vector of each class.Knowledge transfer is achieved by alternately increasing and decreasing the entropy of unlabeled target samples with the classifier and the feature extractor,respectively.Transfer BF fault diagnosis experiments show that our method improves accuracy by about 5%over state-of-the-art methods.

MTTSNet:Military time-sensitive targets stealth network via real-time mask generation

The automatic stealth task of military time-sensitive targets plays a crucial role in maintaining national military security and mastering battlefield dynamics in military applications.We propose a novel Military Time-sensitive Targets Stealth Network via Real-time Mask Generation(MTTSNet).According to our knowledge,this is the first technology to automatically remove military targets in real-time from videos.The critical steps of MTTSNet are as follows:First,we designed a real-time mask generation network based on the encoder-decoder framework,combined with the domain expansion structure,to effectively extract mask images.Specifically,the ASPP structure in the encoder could achieve advanced semantic feature fusion.The decoder stacked high-dimensional information with low-dimensional information to obtain an effective mask layer.Subsequently,the domain expansion module guided the adaptive expansion of mask images.Second,a context adversarial generation network based on gated convolution was constructed to achieve background restoration of mask positions in the original image.In addition,our method worked in an end-to-end manner.A particular semantic segmentation dataset for military time-sensitive targets has been constructed,called the Military Time-sensitive Target Masking Dataset(MTMD).The MTMD dataset experiment successfully demonstrated that this method could create a mask that completely occludes the target and that the target could be hidden in real time using this mask.We demonstrated the concealment performance of our proposed method by comparing it to a number of well-known and highly optimized baselines.

Ultra reliability and massive connectivity provision in integrated internet of military things(IoMT)based on tactical datalink

One of the major challenges arising in internet of military things(IoMT)is accommodating massive connectivity while providing guaranteed quality of service(QoS)in terms of ultra-high reliability.In this regard,this paper presents a class of code-domain nonorthogonal multiple accesses(NOMAs)for uplink ultra reliable networking of massive IoMT based on tactical datalink such as Link-16 and joint tactical information distribution system(JTIDS).In the considered scenario,a satellite equipped with Nr antennas servers K devices including vehicles,drones,ships,sensors,handset radios,etc.Nonorthogonal coded modulation,a special form of multiple input multiple output(MIMO)-NOMA is proposed.The discussion starts with evaluating the output signal to interference-plus-noise(SINR)of receiver filter,leading to the unveiling of a closed-form expression for overloading systems as the number of users is significantly larger than the number of devices admitted such that massive connectivity is rendered.The expression allows for the development of simple yet successful interference suppression based on power allocation and phase shaping techniques that maximizes the sum rate since it is equivalent to fixed-point programming as can be proved.The proposed design is exemplified by nonlinear modulation schemes such as minimum shift keying(MSK)and Gaussian MSK(GMSK),two pivotal modulation formats in IoMT standards such as Link-16 and JITDS.Numerical results show that near capacity performance is offered.Fortunately,the performance is obtained using simple forward error corrections(FECs)of higher coding rate than existing schemes do,while the transmit power is reduced by 6 dB.The proposed design finds wide applications not only in IoMT but also in deep space communications,where ultra reliability and massive connectivity is a keen concern.

基于图的Co-Training网页分类

本文充分利用网页数据的超链接关系和文本信息,提出了一种用于网页分类的归纳式半监督学习算法:基于图的Co-training网页分类算法(Graph based Co-training algorithmfor web page classification),简称GCo-training,并从理论上证明了算法的有效性.GCo-training在Co-training算法框架下,迭代地学习一个基于由超链接信息构造的图的半监督分类器和一个基于文本特征的Bayes分类器.基于图的半监督分类器只利用少量的标记数据,通过挖掘数据间大量的关系信息就可达到比较高的预测精度,可为Bayes分类器提供大量的标记信息;反过来学习大量标记信息后的Bayes分类器也可为基于图的分类器提供有效信息.迭代过程中,二者互相帮助,不断提高各自的性能,而后Bayes分类器可以用来预测大量未见数据的类别.在Web→KB数据集上的实验结果表明,与利用文本特征和锚文本特征的Co-training算法和基于EM的Bayes算法相比,GCo-training算法性能优越.

基于差异性评估对Co-training文本分类算法的改进

Co-training算法要求两个特征视图满足一致性和独立性假设,但是,许多实际应用中不存自然的划分且满足这种假设的两个视图,且直接评估两个视图的独立性有一定的难度.分析Co-training的理论假设,本文把寻找两个满足一致性和独立性特征视图的目标,转变成寻找两个既满足一定的正确性,又存在较大的差异性的两个基分类器的问题.首先利用特征评估函数建立多个特征视图,每个特征视图包含足够的信息训练生成一个基分类器,然后通过评估基分类器之间的差异性间接评估二者的独立性,选择两个满足一定的正确性和差异性比较大的基分类器协同训练.根据每个视图上采用的分类算法是否相同,提出了两种改进算法TV-SC和TV-DC.实验表明改进的TV-SC和TV-DC算法明显优于基于随机分割特征视图的Co-Rnd算法,而且TV-DC算法的分类效果要优于TV-SC算法.

基于Co-training的用户属性预测研究

针对当前基于第三方应用数据进行用户属性预测算法研究,其较少考虑应用前台实际使用时长问题,由此,本文在应用的使用频率及使用时长的基础上,构造了应用前台均使用时长特征,该特征能进一步刻画用户对应用的兴趣度;同时,为充分利用大量未标注数据,从多角度特征对用户属性进行预测,由此本文采用了Co-training框架,该框架包含两个均由栈式自编码器与神经网络相结合的网络结构。实验过程中,对于栈式自编码算法,先利用未标注的数据对网络进行参数初始化,使得网络参数处于一个较优的位置,再利用有标注的数据,采用基于准确率的梯度下降算法,对网络参数进行更新,最终达到收敛。实验结果表明,本文算法在准确率、召回率、F1值上均有所提高。

半监督学习的Co-training算法研究

介绍一种基于半监督学习的协同训练(Co-training)分类算法,当可用的训练样本比较少时,使用传统的方法进行分类,如决策树分类,将无法得到用户满意的结果,而且它们需要大量的标记样本。事实上,获取有标签的样本的代价是相当昂贵的。于是,使用较少的已标记样本和大量的无标记样本进行协同训练的半监督学习,成为研究者首选。

基于样本条件价值改进的Co-training算法

Co-training是一种主流的半监督学习算法.该算法中两视图下的分类器通过迭代的方式,互为对方从无标记样本集中挑选新增样本,以更新对方训练集.Co-training以分类器的后验概率输出作为新增样本的挑选策略,该策略忽略了样本对于当前分类器的价值.针对该问题,本文提出一种改进的Co-training式算法—CVCOT(Conditional value-based co-training),即采用基于样本条件价值的挑选策略来优化Co-training.通过定义无标记样本的条件价值,各视图下的分类器以样本条件价值为依据来挑选新增样本,以此更新训练集.该策略既可保证新增样本的标记可靠性,又能优先将价值较高的富信息样本补充到训练集中,可以有效地优化分类器.在UCI数据集和网页分类应用上的实验结果表明:CVCOT具有较好的分类性能和学习效率.

Co-training机器学习方法在中文组块识别中的应用

采用半指导机器学习方法co training实现中文组块识别。首先明确了中文组块的定义,co training算法的形式化定义。文中提出了基于一致性的co training选取方法将增益的隐马尔可夫模型(TransductiveHMM)和基于转换规则的分类器(fnTBL)组合成一个分类体系,并与自我训练方法进行了比较,在小规模汉语树库语料和大规模未带标汉语语料上进行中文组块识别,实验结果要比单纯使用小规模的树库语料有所提高,F值分别达到了85 34%和83 4 1% ,分别提高了2 13%和7 2 1%。

用于在线产品评论质量分析的Co-training算法

在线评论广泛存在于电子商务网站平台,其中包含着客户对产品的评价及偏好.高效分析在线评论数据并满足客户需求,对许多谋求立足于竞争激烈的国际化市场的企业来说至关重要.但因在线评论的质量不一,使得如何分析在线评论的质量成为一项重要工作.从两个方面提取特征对在线评论进行描述,并构建了一种Co-training算法来判断评论的质量.通过对比实验验证了该算法相对于单一分类算法的优势.

基于Co-training训练CRF模型的评价对象识别

评价对象是指某段评论中评价词语所修饰的对象或对象的属性。为了识别评论中的评价对象,提出基于Co-training的训练CRF模型方法。该方法首先人工标注少量的原始数据集,使用Co-training方式对未标注数据进行自动识别,以扩大已标注训练数据。通过原始标注数据集和Co-training方式标注数据集,训练CRF模型。在汽车领域中,对待标注汽车评论语料中评价对象识别的精确率为67.483%,召回率为67.832%。

Feature selection for co-training

Co-training is a semi-supervised learning method, which employs two complementary learners to label the unlabeled data for each other and to predict the test sample together. Previous studies show that redundant information can help improve the ratio of prediction accuracy between semi-supervised learning methods and supervised learning methods. However, redundant information often practically hurts the performance of learning machines. This paper investigates what redundant features have effect on the semi-supervised learning methods, e.g. co-training, and how to remove the redundant features as well as the irrelevant features. Here, FESCOT （feature selection for co-training） is proposed to improve the generalization performance of co-training with feature selection. Experimental results on artificial and real world data sets show that FESCOT helps to remove irrelevant and redundant features that hurt the performance of the co-training method.

Semi-supervised LIBS quantitative analysis method based on co-training regression model with selection of effective unlabeled samples

The accuracy of laser-induced breakdown spectroscopy(LIBS) quantitative method is greatly dependent on the amount of certified standard samples used for training. However, in practical applications, only limited standard samples with labeled certified concentrations are available. A novel semi-supervised LIBS quantitative analysis method is proposed, based on co-training regression model with selection of effective unlabeled samples. The main idea of the proposed method is to obtain better regression performance by adding effective unlabeled samples in semisupervised learning. First, effective unlabeled samples are selected according to the testing samples by Euclidean metric. Two original regression models based on least squares support vector machine with different parameters are trained by the labeled samples separately, and then the effective unlabeled samples predicted by the two models are used to enlarge the training dataset based on labeling confidence estimation. The final predictions of the proposed method on the testing samples will be determined by weighted combinations of the predictions of two updated regression models. Chromium concentration analysis experiments of 23 certified standard high-alloy steel samples were carried out, in which 5 samples with labeled concentrations and 11 unlabeled samples were used to train the regression models and the remaining 7 samples were used for testing. With the numbers of effective unlabeled samples increasing, the root mean square error of the proposed method went down from 1.80% to 0.84% and the relative prediction error was reduced from 9.15% to 4.04%.

Pushing the frontiers of military medical excellence:updates,progress and future needs

Since its establishment in 2014,Military Medical Research has come a long way in becoming a premier journal for scientific articles from various different specialties,with a special emphasis on topics with military relevance.The field of military medicine may be obscure,and may not be readily encountered by the typical clinician on a day-today basis.This journal aims not only to pursue excellence in military research,but also to keep current with the latest advancements on general medical topics from each and every specialty.This editorial serves to recap and synthesize the existing progress,updates and future needs of military medical excellence,discussing foremostly the unique traits of literature published in this journal,and subsequently presenting the discourse regarding wartime and peacetime medicine,the role of the military in a public health emergency,as well as wound healing and organ regeneration.Special attention has been devoted to military topics to shed light on the effects of Chemical,Biological,Radiological and Explosive warfare,environmental medicine and military psychiatry,topics which rarely have a chance to be discussed elsewhere.The interconnectedness between military combat and soldier physical and mental well-being is intricate,and has been distorted by pandemics such as coronavirus disease 2019(COVID-19).This journal has come a long way since its first article was published,steadily contributing to the existing knowledge pool on general medical topics with a military slant.Only with continuous research and sharing,can we build upon the work of the scientific community,with hopes for the betterment of patient care.

Research of vibrations effect on hydraulic valves in military vehicles

The paper discusses minimizing the effect of external mechanical vibration on hydraulic valves in different military hydraulic drive systems.The current research work presents an analysis of the potential to reduce vibration on the valve casing by installing a valve flexibly on a vibrating surface,i.e.,by introducing a material with known stiffness and damping characteristics between the valve casing and the vibrating surface-a steel spring package or special cushions made of elastomer material or of oilresistant rubber.The article also demonstrates that elastomer cushions placed inside the valve casingbetween the casing and the centering springs-can be used as a supplementary or alternative solution in the analyzed method for mitigating the transfer of vibrations.By using materials with appropriately selected elastic and dissipative properties,the effectiveness of vibro-isolation can be increased.The presented theoretical analyzes by linear and non-linear mathematical models have been verified experimentally.

Lethal means and suicide prevention among military veterans

Dear Editor,Suicide amongst the military veteran population is a significant publichealthproblemintheUnitedStates.TheNational VeteranSuicidePreventionAnnualReportrevealedthat6261 died by suicide in 2019[1]. The lingering effects of the coronavirus disease 2019 (COVID-19) pandemic may account for an increase in veteran suicide rates[1].

Could Military Commanders’ Good LeadershipInfluence Subordinates’ Smartphone Overdependence? A Serial Mediation Analysis

Owing to the ubiquitous use of smartphones by soldiers, military researchers have an increasing interest in potentially problematicside effects such as smartphone overdependence. This raises a question regarding the psychological mechanisms underlying thepotentially self-damaging use of smartphones. Here, we address this question by analyzing how heterogeneity in commander’sgood leadership explains subordinate soldiers’ differences in self-control and smartphone use. Specifically, we found thatsubordinate soldiers who thought their commander's leadership was good were self-regulated, less dependent on smartphones,less stressed, and finally had good mental health. This result indicates that commander’s good leadership can be used toestimate whether subordinate soldiers exert control over their impulses and use their smartphones properly. Thus, the currentfindings help to identify external factors that lead to a better understanding of problematic smartphone use and can potentiallyhelp to design appropriate preventive mechanisms or interventions that target commander’s good leadership.

Epidemio-Clinical Profile of Psychotrauma in Military Victims of Terrorists Attacks in Benin in December 2021

Introduction: In December 2021, Beninese soldiers were victims of armed terrorist attacks. This study aims to determine the psychological impact of the events experienced on the victims in order to better prevent post-traumatic stress disorder. Materials and methods: This is a descriptive cross-sectional study conducted from December 13 to 14, 2021, i.e. 12 days after the oldest event and 4 days after the most recent. The experience of the event was assessed with the Peritraumatic Distress Inventory and the Peritraumatic Dissociation Experiences Questionnaire. Symptoms experienced in the hours and days after the event were assessed with the Revised Event Impact Scale. Results: 36 soldiers were included out of the 38 survivors, i.e. 94.73%. They were 51.43% victims of the night assault, 37.14% of the ambush on patrol and 11.43% of the explosion of vehicle on mine, all male. The average age was 31 years with extremes of 25 and 49 years. The 25 - 30 year olds predominated, 52.78%. 100% of soldiers had significant peritraumatic dissociation, of which 94.44% also had significant peritraumatic distress. 100% of the soldiers had manifestations of intrusion, 89% of manifestations of avoidance and 83% of manifestations of hyperexcitation. 88.89% had an index for acute stress including 100% of victims of night assault and vehicle explosion on mine. Conclusion: This study shows a high prevalence of peritraumatic reactions, recognized factors favoring post-traumatic stress disorder, and the interest of their evaluation for medical and psychological help adapted to the victims.

Masquelet technique in military practice:specificities and future directions for combat-related bone defect reconstruction

Because of its simplicity,reliability,and replicability,the Masquelet induced membrane technique(IMT)has become one of the preferred methods for critical bone defect reconstruction in extremities.Although it is now used worldwide,few studies have been published about IMT in military practice.Bone reconstruction is particularly challenging in this context of care due to extensive soft-tissue injury,early wound infection,and even delayed management in austere conditions.Based on our clinical expertise,recent research,and a literature analysis,this narrative review provides an overview of the IMT application to combat-related bone defects.It presents technical specificities and future developments aiming to optimize IMT outcomes,including for the management of massive multi-tissue defects or bone reconstruction performed in the field with limited resources.

High risk for obstructive sleep apnea and risk of hypertension in military personnel: The CHIEF sleep study

BACKGROUND Epidemiological studies have revealed an association between obstructive sleep apnea(OSA)and hypertension in the general population,while the association in military personnel was rarely investigated.AIM To examine the association between high risk for OSA and hypertension by phenotypes in military young adults.METHODS A total of 746 military personnel,aged 27.9 years,were included in the cardiorespiratory fitness and health in armed forces(CHIEF)-sleep study in Taiwan in 2020.Antihypertensive medications were not used by the subjects.High risk for OSA was assessed using the Berlin Questionnaire.Hypertension was defined using the 7th Report of the Joint National Committee on Prevention,Detection,Evaluation,and Treatment of High Blood Pressure(JNC 7)and the 2017 American College of Cardiology(ACC)/American Heart Association(AHA)guidelines.The cutoff levels of systolic and diastolic blood pressure(SBP and DBP)for the 2017 ACC/AHA-and JNC 7-based guidelines were 130/140 mmHg and 80/90 mmHg,respectively.Hypertension phenotypes included isolated systolic and diastolic hypertension(ISH,high SBP only and IDH,high DBP only)and combined hypertension(both high SBP and DBP).Multivariable logistic regression analysis with adjustment for demographics,lifestyle and metabolic biomarkers.RESULTS The prevalence of high risk for OSA,JNC 7-based hypertension and 2017 ACC/AHA-based hypertension were 8.0%,5.2%and 22.0%,respectively.Those with a high risk for OSA had a higher probability of JNC 7-based overall and combined hypertension(odds ratios(ORs))and 95%confidence intervals:2.82(1.07-7.42)and 7.54(1.10-51.54),although the probabilities of ISH and IDH were unaffected by a high risk for OSA(ORs:1.96 and 2.35,respectively,both P>0.05).In contrast,no associations for any hypertension phenotypes were found according to the 2017 ACC/AHA criteria.CONCLUSION A high risk for OSA was associated with severe hypertension and combined hypertension among Asian military young adults.