Performance Evaluation Indicator (PEI): A new paradigm to evaluate the competence of machine learning classifiers in predicting rockmass conditions

To illustrate an unprejudiced comparison among machine learning classifiers established on proprietary databases, and to guarantee the validity and robustness of these classifiers, a Performance Evaluation Indicator (PEI) and the corresponding failure criterion are proposed in this study. Three types of machine learning classifiers, including the strictly binary classifier, the normal multiclass classifier and the misclassification cost-sensitive classifier, are trained on four datasets recorded from a water drainage TBM project. The results indicate that: (1) the PEI successfully compares the competence of classifiers under different scenarios by isolating the effects of different overlapping-degree of rockmass classes, and (2) the cost-sensitive algorithm is warranted to classify rockmasses when the ratio of inter-class classes is more than 8:1. The contributions of this research are to fill the gap in performance evaluations of a classifier for imbalanced training data, and to identify the best situation to apply this classifier.     

Dynamic ensemble wind speed prediction model based on hybrid deep reinforcement learning

Prediction of wind speed can provide a reference for the reliable utilization of wind energy. This study focuses on 1-hour, 1-step ahead deterministic wind speed prediction with only wind speed as input. To consider the time-varying characteristics of wind speed series, a dynamic ensemble wind speed prediction model based on deep reinforcement learning is proposed. It includes ensemble learning, multi-objective optimization, and deep reinforcement learning to ensure effectiveness. In part A, deep echo state network enhanced by real-time wavelet packet decomposition is used to construct base models with different vanishing moments. The variety of vanishing moments naturally guarantees the diversity of base models. In part B, multi-objective optimization is adopted to determine the combination weights of base models. The bias and variance of ensemble model are synchronously minimized to improve generalization ability. In part C, the non-dominated solutions of combination weights are embedded into a deep reinforcement learning environment to achieve dynamic selection. By reasonably designing the reinforcement learning environment, it can dynamically select non-dominated solution in each prediction according to the time-varying characteristics of wind speed. Four actual wind speed series are used to validate the proposed dynamic ensemble model. The results show that: (a) The proposed dynamic ensemble model is competitive for wind speed prediction. It significantly outperforms five classic intelligent prediction models and six ensemble methods; (b) Every part of the proposed model is indispensable to improve the prediction accuracy.     

Artificial intelligence for BIM content management and delivery: Case study of association rule mining for construction detailing

The proliferation of Building Information Modeling (BIM) applications, in tandem with the extensive variation of building products, pose new demands on design and engineering firms to efficiently manage and reuse BIM content (i.e., data-rich parametric model objects and assembly details). Tasks such as classifying BIM objects, indexing them with meta-data (e.g., category), and searching digital libraries to load objects into models still plague practice with inefficient manual workflows. This research aims to improve the productivity of BIM content management and retrieval by developing an AI-backed BIM content recommender system. Using data from a case-study firm, this research extracted content from over 30,000 technical BIM views (e.g., plans, sections, details) in historical projects to build an unsupervised machine-learning prototype with association rule mining. This prototype explicated the strength of relationships among co-occurring BIM objects. Using this prototype as the backbone AI-engine in live BIM sessions, this research developed a context-aware recommender system that dynamically provides BIM users with a set of objects associable with their modeling context (e.g., type of view, existing objects in the model) and human–computer interactions (e.g., objects selected by the user). By mining association data from hundreds of historical projects, this development marks a departure from the existing prototypes that rely on explicit coding, recurring user input, or subjective ratings to recommend BIM content to users. The simulation and experimental implementation of this recommender system yielded high efficacy in predicting content needs and achieved significant savings in the time spent on conventional BIM workflows.     

A novel deep learning prediction model for concrete dam displacements using interpretable mixed attention mechanism

Dam displacements can effectively reflect its operational status, and thus establishing a reliable displacement prediction model is important for dam health monitoring. The majority of the existing data-driven models, however, focus on static regression relationships, which cannot capture the long-term temporal dependencies and adaptively select the most relevant influencing factors to perform predictions. Moreover, the emerging modeling tools such as machine learning (ML) and deep learning (DL) are mostly black-box models, which makes their physical interpretation challenging and greatly limits their practical engineering applications. To address these issues, this paper proposes an interpretable mixed attention mechanism long short-term memory (MAM-LSTM) model based on an encoder-decoder architecture, which is formulated in two stages. In the encoder stage, a factor attention mechanism is developed to adaptively select the highly influential factors at each time step by referring to the previous hidden state. In the decoder stage, a temporal attention mechanism is introduced to properly extract the key time segments by identifying the relevant hidden states across all the time steps. For interpretation purpose, our emphasis is placed on the quantification and visualization of factor and temporal attention weights. Finally, the effectiveness of the proposed model is verified using monitoring data collected from a real-world dam, where its accuracy is compared to a classical statistical model, conventional ML models, and homogeneous DL models. The comparison demonstrates that the MAM-LSTM model outperforms the other models in most cases. Furthermore, the interpretation of global attention weights confirms the physical rationality of our attention-based model. This work addresses the research gap in interpretable artificial intelligence for dam displacement prediction and delivers a model with both high-accuracy and interpretability.     

A joint introduction to Gaussian Processes and Relevance Vector Machines with connections to Kalman filtering and other kernel smoothers

The expressive power of Bayesian kernel-based methods has led them to become an important tool across many different facets of artificial intelligence, and useful to a plethora of modern application domains, providing both power and interpretability via uncertainty analysis. This article introduces and discusses two methods which straddle the areas of probabilistic Bayesian schemes and kernel methods for regression: Gaussian Processes and Relevance Vector Machines. Our focus is on developing a common framework with which to view these methods, via intermediate methods a probabilistic version of the well-known kernel ridge regression, and drawing connections among them, via dual formulations, and discussion of their application in the context of major tasks: regression, smoothing, interpolation, and filtering. Overall, we provide understanding of the mathematical concepts behind these models, and we summarize and discuss in depth different interpretations and highlight the relationship to other methods, such as linear kernel smoothers, Kalman filtering and Fourier approximations. Throughout, we provide numerous figures to promote understanding, and we make numerous recommendations to practitioners. Benefits and drawbacks of the different techniques are highlighted. To our knowledge, this is the most in-depth study of its kind to date focused on these two methods, and will be relevant to theoretical understanding and practitioners throughout the domains of data-science, signal processing, machine learning, and artificial intelligence in general.     

基于信息传播节点集的CTDN节点分类算法

针对连续时间动态网络的节点分类问题,根据实际网络信息传播特点定义信息传播节点集,改进网络表示学习的节点序列采样策略,并设计基于信息传播节点集的连续时间动态网络节点分类算法,通过网络表示学习方法生成的节点低维向量以及OpenNE框架内的LogicRegression分类器,获得连续时间动态网络的节点分类结果。实验结果表明,与CTDNE和STWalk算法相比,该算法在实验条件相同的情况下,网络表示学习结果的二维可视化效果更优且最终的网络节点分类精度更高。     

基于显著性多尺度特征协作融合的行人重识别方法

由于背景信息复杂、遮挡等因素的影响,现有基于局部特征的行人重识别方法所提取的特征不具有辨别力和鲁棒性,从而导致重识别精度较低,针对该问题,提出一种基于显著性检测与多尺度特征协作融合的SMC-ReID方法。利用显著性检测提取行人中具有判别力的特征区域,融合显著性特征与全局特征并完成不同尺度的切块,将上述不同尺度的特征进行协作融合以保证特征切块后的连续性,根据全局特征和局部特征的差异性联合3种损失函数进行学习。在推理阶段,将各个尺度的特征降低到同一维度并融合成新的特征向量,以实现相似性度量。在行人重识别公开数据集Market1501、DukeMTMC-reID和CUHK03上进行实验,结果表明,SMC-ReID方法所提取的特征具有较强的可区分性和鲁棒性,识别准确率优于SVDNet和PSE+ECN等方法。     

一种基于对抗学习的实时跟踪模型设计

目标跟踪指在视频帧中找到感兴趣目标的运动位置,广泛应用于环境感知、安防监控和无人驾驶等领域。为进行高效的目标跟踪,建立一种基于对抗学习和特征压缩的相关滤波器目标跟踪模型。为了同时兼顾精度与速度,在模型中引入特征提取优化、特征压缩和特征聚合等步骤。在提取图像特征前,采用对抗学习方法解决特征提取模型中训练数据与任务数据分布不匹配的问题。在特征压缩阶段,应用双通道自编码器结构和特征聚合来增强模型对图像风格的泛化能力。实验结果表明,与非实时跟踪算法相比,该模型在精度损失不超过3%的情况下能取得明显的速度提升,其跟踪速度高达103FPS。     

基于深度学习的内容推荐算法研究综述

推荐系统是学习用户偏好,实现个性化推荐的系统化应用技术,在商品购买、影音推荐、关联阅读等多领域得到了广泛的应用。近年来,随着多源异构数据的激增和深度学习的兴起,传统推荐算法中的表征学习模式逐步被深度学习代替。梳理推荐算法的背景和发展趋势,并给出内容推荐的算法思路及其优劣评价,分别介绍多层感知机、自动编码器、卷积神经网络以及循环神经网络等深度学习方法的网络结构和算法优势。从技术应用的视角综述深度学习在内容推荐中的应用现状与研究成果,对不同经典深度推荐算法进行分析与比较。在此基础上,指出深度学习在可解释性、学习效率等方面的不足,并对交叉领域学习、多任务学习、表征学习等未来研究方向进行展望。     

Mask R-CNN中特征不平衡问题的全局信息融合方法

特征不平衡问题是影响神经网络检测效率的关键因素。针对Mask R-CNN中的特征不平衡问题,提出一种基于全局特征金字塔网络（GFPN）的信息融合方法。通过将GFPN产生的不同大小特征相融合,生成包含全局语义信息的特征网络,并采用反向过程对原始特征层进行重新标度,从而使得每个特征层均含有全局语义信息。实验结果表明,与原始基于Mask R-CNN的方法相比,该方法的检测精度提升4~6个百分点,而检测时间仅增加0.112 s。