Convolutional neural network: Deep learning-based classification of building quality problems

The rapid development of the construction industry in China has introduced unprecedented quality-related problems in the country’s building industry. In response to this issue, the government has established various complaint channels to report quality problems. Therefore, building quality complaints (BQCs) need to be classified and solved by respective agencies or departments rapidly for avoiding adverse impact on the safety, health, and well-being of people. However, the current process of classifying BQCs is labor intensive, time consuming, and error prone. An automatic complaint classification is required to improve the effectiveness and efficiency of complaint handling, but studies on this issue are limited. Prevailing text classification research in construction has focused on utilizing conventional shallow machine learning. By contrast, this study explores a novel convolutional neural network (CNN)-based approach that incorporates a deep-learning method to automatically classify the short texts contained within BQCs. The presented approach enables capturing the semantic features in BQC texts and automatic classification of the BQCs into predefined categories. After the model optimization, tests are conducted to examine the practical application of the text classification approach compared with Bayes-based and support vector machine classifiers. Results indicate that the developed CNN-based approach performs well in the Chinese BQC classification with limited manual intervention and few complicated feature engineering.     

基于优势学习的深度Q网络

强化学习问题中，同一状态下不同动作所对应的状态-动作值存在差距过小的现象，Q-Learning算法采用MAX进行动作选择时会出现过估计问题，且结合了Q-Learning的深度Q网络（Deep Q Net）同样存在过估计问题。为了缓解深度Q网络中存在的过估计问题，提出一种基于优势学习的深度Q网络，通过优势学习的方法构造一个更正项，利用目标值网络对更正项进行建模，同时与深度Q网络的评估函数进行求和作为新的评估函数。当选择的动作是最优动作时，更正项为零，不对评估函数的值进行改动，当选择的动作不是最优动作时，更正项的值为负，降低了非最优动作的评估值。和传统的深度Q网络相比，基于优势学习的深度Q网络在Playing Atari 2600的控制问题breakout、seaquest、phoenix、amidar中取得了更高的平均奖赏值，在krull、seaquest中取得了更加稳定的策略。     

结合深度学习与支持向量机的金属零件识别

目的 在视觉引导的工业机器人自动拾取研究中，关键技术难点之一是机器人抓取目标区域的识别问题。特别是金属零件，其表面的反光、随意摆放时相互遮挡等非结构化因素都给抓取区域的识别带来巨大的挑战。因此，本文提出一种结合深度学习和支持向量机的抓取区域识别方法。方法 分别提取抓取区域的方向梯度直方图（HOG）和局部二进制模式（LBP）特征，利用主成分分析法（PCA）对融合后的特征进行降维，以此来训练支持向量机（SVM）分类器。通过训练Mask R-CNN（regions with convolutional neural network）神经网络完成抓取区域的初步分割。然后利用SVM对Mask R-CNN识别的抓取区域进行二次分类，完成对干扰区域的剔除。最后计算掩码完成实例分割，以此达到对抓取区域的精确识别。结果 对于随机摆放的铜质金属零件，本文算法与单一的Mask R-CNN及多特征融合的SVM算法就识别准确率、错检率、漏检率3个指标进行了比较，结果表明本文算法在识别准确率上较Mask R-CNN和SVM算法分别提高了7%和25%，同时有效降低了错检率与漏检率。结论 本文算法结合了Mask R-CNN与SVM两种方法，对于反光和遮挡情况具有一定的鲁棒性，同时有效地提升了目标识别的准确率。     

An engineering method for complex structural optimization involving both size and topology design variables

This work presents an engineering method for optimizing structures made of bars, beams, plates, or a combination of those components. Corresponding problems involve both continuous (size) and discrete (topology) variables. Using a branched multipoint approximate function, which involves such mixed variables, a series of sequential approximate problems are constructed to make the primal problem explicit. To solve the approximate problems, genetic algorithm (GA) is utilized to optimize discrete variables, and when calculating individual fitness values in GA, a second-level approximate problem only involving retained continuous variables is built to optimize continuous variables. The solution to the second-level approximate problem can be easily obtained with dual methods. Structural analyses are only needed before improving the branched approximate functions in the iteration cycles. The method aims at optimal design of discrete structures consisting of bars, beams, plates, or other components. Numerical examples are given to illustrate its effectiveness, including frame topology optimization, layout optimization of stiffeners modeled with beams or shells, concurrent layout optimization of beam and shell components, and an application in a microsatellite structure. Optimization results show that the number of structural analyses is dramatically decreased when compared with pure GA while even comparable to pure sizing optimization.     

Multiple Empirical Kernel Learning with Majority Projection for imbalanced problems

Traditional Multiple Empirical Kernel Learning (MEKL) expands the expressions of the sample and brings better classification ability by using different empirical kernels to map the original data space into multiple kernel spaces. To make MEKL suit for the imbalanced problems, this paper introduces a weight matrix and a regularization term into MEKL. The weight matrix assigns high misclassification cost to the minority samples to balanced misclassification cost between minority and majority class. The regularization term named Majority Projection (MP) is used to make the classification hyperplane fit the distribution shape of majority samples and enlarge the between-class distance of minority and majority class. The contributions of this work are: (i) assigning high cost to minority samples to deal with imbalanced problems, (ii) introducing a new regularization term to concern the property of data distribution, (iii) and modifying the original PAC-Bayes bound to test the error upper bound of MEKL-MP. Through analyzing the experimental results, the proposed MEKL-MP is well suited to the imbalanced problems and has lower generalization risk in accordance with the value of PAC-Bayes bound.     

Dialogic weaving: a favorable tension between design and craft

Focusing on a process where a designer embeds herself in craftspeople’s workshops with the intention of ‘learning from inside’ and ‘making together’; this paper proposes pathways for considering collaboration between designers and craftspeople via digital fabrication and weaving. The tension of forces that make weaving possible, act as an analogy for this collaboration as well as providing the title of the project as ‘One Over, One Under’. The project proposes a mode of working and a spectrum of outcomes where the designer has a first-hand experience of production techniques, engages in a serious and continuous dialogue through making, and develops an intervention that also brings forth her own skill set (introducing digital fabrication and structural variations), thus transforming the conventions and boundaries between established roles and manufacturing techniques. An experiment that resulted in a series of objects considering both the technologies of production and the input of the designer, this process not only increases the potentials that crafts hold for the field of design, but it also offers possibilities of collaboration and a further articulation of the design act.     

Nonuniform linear array DOA estimation using EM criterion

In this letter, we address the problem of Direction of Arrival (DOA) estimation with nonuniform linear array in the context of sparse Bayesian learning (SBL) framework. The nonuniform array output is deemed as an incomplete-data observation, and a hypothetical uniform linear array output is treated as an unavailable complete-data observation. Then the Expectation-Maximization (EM) criterion is directly utilized to iteratively maximize the expected value of the complete-data log likelihood under the posterior distribution of the latent variable. The novelties of the proposed method lie in its capability of interpolating the actual received data to a virtual uniform linear array, therefore extending the achievable array aperture. Simulation results manifests the superiority of the proposed method over off-the-shelf algorithms, specially on circumstances such as low SNR, insufficient snapshots, and spatially adjacent sources.     

基于深度学习的视频跟踪研究进展综述

近年来深度学习迅猛发展，颠覆了语音识别、图像分类、文本理解等领域的算法设计思路。深度学习因其具备强大的特征提取能力，在图像识别领域的成绩尤为突出。然而深度学习与视频监控领域的结合并不多，由于深度模型具有多层网络结构，算法复杂度大，训练和更新模型时比较耗时，很难满足实时性要求。回顾了深度学习的发展史，介绍了最近10年来国内外深度学习主要模型，论述了基于深度学习的目标跟踪算法，指出了各算法的优缺点，最后对当前该领域存在的问题和发展前景进行了总结和展望。     

基于聚类的数据加权优化在犯罪预测中的应用

近年来，我国传统暴力犯罪与成年人犯罪呈下降态势，但是，犯罪案由层出不穷。为有效提升公安实践工作中犯罪预测能力，打击各类违法犯罪事件,本文针对犯罪数据，提出一种新型犯罪预测模型。利用密度聚类分析方法将犯罪数据分类，然后进行数据降维提取关键属性生成特征数据，继而对特征数据进行加权优化并采用机器学习的方式对特征数据进行学习，从而预测犯罪案由。实验结果表明，与传统方法相比，本文方法具有更好的预测效果，为公安实践工作中类似案件的侦破和预防，提供新的路径支撑。