Augmented Industrial Data-Driven Modeling Under the Curse of Dimensionality

The curse of dimensionality refers to the problem of increased sparsity and computational complexity when dealing with high-dimensional data. In recent years, the types and variables of industrial data have increased significantly, making data-driven models more challenging to develop. To address this problem, data augmentation technology has been introduced as an effective tool to solve the sparsity problem of high-dimensional industrial data. This paper systematically explores and discusses the necessity, feasibility, and effectiveness of augmented industrial data-driven modeling in the context of the curse of dimensionality and virtual big data. Then, the process of data augmentation modeling is analyzed, and the concept of data boosting augmentation is proposed. The data boosting augmentation involves designing the reliability weight and actual-virtual weight functions, and developing a double weighted partial least squares model to optimize the three stages of data generation, data fusion, and modeling. This approach significantly improves the interpretability, effectiveness, and practicality of data augmentation in the industrial modeling. Finally, the proposed method is verified using practical examples of fault diagnosis systems and virtual measurement systems in the industry. The results demonstrate the effectiveness of the proposed approach in improving the accuracy and robustness of data-driven models, making them more suitable for real-world industrial applications.      

深度学习模型可解释性研究综述

深度学习技术以数据驱动学习的特点,在自然语言处理、图像处理、语音识别等领域取得了巨大成就。但由于深度学习模型网络过深、参数多、复杂度高等特性,该模型做出的决策及中间过程让人类难以理解,因此探究深度学习的可解释性成为当前人工智能领域研究的新课题。以深度学习模型可解释性为研究对象,对其研究进展进行总结阐述。从自解释模型、特定模型解释、不可知模型解释、因果可解释性四个方面对主要可解释性方法进行总结分析。列举出可解释性相关技术的应用,讨论当前可解释性研究存在的问题并进行展望,以推动深度学习可解释性研究框架的进一步发展。     

AutoConfig: 面向深度学习编译优化的自动配置机制

随着深度学习模型和硬件架构的快速发展,深度学习编译器已经被广泛应用.目前,深度学习模型的编译优化和调优的方法主要依赖基于高性能算子库的手动调优和基于搜索的自动调优策略.然而,面对多变的目标算子和多种硬件平台的适配需求,高性能算子库往往需要为各种架构进行多次重复实现.此外,现有的自动调优方案也面临着搜索开销大和缺乏可解释性的挑战.为了解决上述问题,本文提出了AutoConfig,一种面向深度学习编译优化的自动配置机制.针对不同的深度学习计算负载和特定的硬件平台,AutoConfig可以构建具备可解释性的优化算法分析模型,采用静态信息提取和动态开销测量的方法进行综合分析,并基于分析结果利用可配置的代码生成技术自动完成算法选择和调优.本文创新性地将优化分析模型与可配置的代码生成策略相结合,不仅保证了性能加速效果,还减少了重复开发的开销,同时简化了调优过程.在此基础上,本文进一步将AutoConfig集成到深度学习编译器Buddy Compiler中,对矩阵乘法和卷积的多种优化算法建立分析模型,并将自动配置的代码生成策略应用在多种SIMD硬件平台上进行评估.实验结果验证了AutoConfig在代码生成策略中有效地完成了参数配置和算法选择.与经过手动或自动优化的代码相比,由AutoConfig生成的代码可达到相似的执行性能,并且无需承担手动调优的重复实现开销和自动调优的搜索开销.     

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.     

面向图像分类的深度模型可解释性研究综述

深度学习目前在计算机视觉、自然语言处理、语音识别等领域得到了深入发展,与传统的机器学习算法相比,深度模型在许多任务上具有较高的准确率.然而,作为端到端的具有高度非线性的复杂模型,深度模型的可解释性没有传统机器学习算法好,这为深度学习在现实生活中的应用带来了一定的阻碍.深度模型的可解释性研究具有重大意义而且是非常必要的,近年来许多学者围绕这一问题提出了不同的算法.针对图像分类任务,将可解释性算法分为全局可解释性和局部可解释性算法.在解释的粒度上,进一步将全局解释性算法分为模型级和神经元级的可解释性算法,将局部可解释性算法划分为像素级特征、概念级特征以及图像级特征可解释性算法.基于上述分类框架,总结了常见的深度模型可解释性算法以及相关的评价指标,同时讨论了可解释性研究面临的挑战和未来的研究方向.认为深度模型的可解释性研究和理论基础研究是打开深度模型黑箱的必要途径,同时可解释性算法存在巨大潜力可以为解决深度模型的公平性、泛化性等其他问题提供帮助.     

深度学习模型可解释性的研究进展

深度学习在很多人工智能应用领域中取得成功的关键原因在于,通过复杂的深层网络模型从海量数据中学习丰富的知识。然而,深度学习模型内部高度的复杂性常导致人们难以理解模型的决策结果,造成深度学习模型的不可解释性,从而限制了模型的实际部署。因此,亟需提高深度学习模型的可解释性,使模型透明化,以推动人工智能领域研究的发展。本文旨在对深度学习模型可解释性的研究进展进行系统性的调研,从可解释性原理的角度对现有方法进行分类,并且结合可解释性方法在人工智能领域的实际应用,分析目前可解释性研究存在的问题,以及深度学习模型可解释性的发展趋势。为全面掌握模型可解释性的研究进展以及未来的研究方向提供新的思路。     

Latent variable model predictive control for trajectory tracking in batch processes: Alternative modeling approaches

Several Latent Variable Model (LVM) structures for modeling the time histories of batch processes are investigated from the view point of their suitability for use in Latent Variable Model Predictive Control (LV-MPC) [1] for trajectory tracking and disturbance rejection in batch processes. The LVMs are based on Principal Component Analysis (PCA). Two previously proposed approaches (Batch-Wise Unfolding (BWU) and Observation-Wise with Time-lag Unfolding (OWTU)) for modeling of batch processes [2] are incorporated in the LV-MPC and the benefits and drawbacks of each are explored. Furthermore, a new modeling approach (Regularized Batch-Wise Unfolding (RBWU)) is proposed to overcome the shortcomings of each of the previous modeling approaches while keeping the major benefits of both. The performances of the three latent variable modeling approaches in the course of LV-MPC for trajectory tracking and disturbance rejection are illustrated using two simulated batch reactor case studies. It is seen that the RBWU approach models the nonlinearity and time-varying properties of the batch almost as accurately as BWU approach, but needs fewer observations (batches) for model identification and results in a smoother PCA model. Recommendations are then given on which modeling approach to use under different scenarios.     

多源空—谱遥感图像融合方法进展与挑战

多源空—谱遥感图像融合方法作为两路不完全观测多通道数据的计算重构反问题,其挑战在于补充信息不足、模糊和噪声等引起的病态性,现有方法在互补特征保持的分辨率增强方面仍有很大的改进空间。为了推动遥感图像融合技术的发展,本文系统概述目前融合建模的代表性方法,包括成分替代、多分辨率分析、变量回归、贝叶斯、变分和模型与数据混合驱动等方法体系及其存在问题。从贝叶斯融合建模的角度,分析了互补特征保真和图像先验在优化融合中的关键作用和建模机理,并综述了目前若干图像先验建模的新趋势,包括：分数阶正则化、非局部正则化、结构化稀疏表示、矩阵低秩至张量低秩表示、解析先验与深度先验的复合等。本文对各领域面临的主要挑战和可能的研究方向进行了概述和讨论,指出解析模型和数据混合驱动将是图像融合的重要发展方向,并需要结合成像退化机理、数据紧致表示和高效计算等问题,突破现有模型优化融合的技术瓶颈,进一步发展更优良的光谱信息保真和更低算法复杂度的融合方法。同时,为了解决大数据问题,有必要在Hadoop和SPARK等大数据平台上进行高性能计算,以更有利于多源数据融合算法的加速实现。     

深度学习可解释性研究综述

随着深度学习的广泛应用,人类越来越依赖于大量采用深度学习技术的复杂系统,然而,深度学习模型的黑盒特性对其在关键任务应用中的使用提出了挑战,引发了道德和法律方面的担忧,因此,使深度学习模型具有可解释性是使它们令人信服首先要解决的问题。于是,关于可解释的人工智能领域的研究应运而生,主要集中于向人类观察者明确解释模型的决策或行为。对深度学习可解释性的研究现状进行综述,为进一步深入研究建立更高效且具有可解释性的深度学习模型确立良好的基础。首先,对深度学习可解释性进行了概述,阐明可解释性研究的需求和定义;然后,从解释深度学习模型的逻辑规则、决策归因和内部结构表示这三个方面出发介绍了几种可解释性研究的典型模型和算法,另外还指出了三种常见的内置可解释模型的构建方法;最后,简单介绍了忠实度、准确性、鲁棒性和可理解性这四种评价指标,并讨论了深度学习可解释性未来可能的发展方向。     

注意力机制在深度学习中的研究进展

注意力机制逐渐成为目前深度学习领域的主流方法和研究热点之一,它通过改进源语言表达方式,在解码中动态选择源语言相关信息,从而极大改善了经典Encoder-Decoder框架的不足。该文在提出传统基于Encoder-Decoder框架中存在的长程记忆能力有限、序列转化过程中的相互关系、模型动态结构输出质量等问题的基础上,描述了注意力机制的定义和原理,介绍了多种不同的分类方式,分析了目前的研究现状,并叙述了目前注意力机制在图像识别、语音识别和自然语言处理等重要领域的应用情况。同时,进一步从多模态注意力机制、注意力的评价机制、模型的可解释性及注意力与新模型的融合等方面进行了探讨,从而为注意力机制在深度学习中的应用提供新的研究线索与方向。     

Data-driven symbolic ensemble models for wind speed forecasting through evolutionary algorithms

Non-linear data-driven symbolic models have been gaining traction in many fields due to their distinctive combination of modeling expressiveness and interpretability. Despite that, they are still rather unexplored for ensemble wind speed forecasting, leaving behind new promising avenues for advancing the development of more accurate models which impact the efficiency of energy production. In this work, we develop a methodology based on the evolutionary algorithm known as grammatical evolution, and apply it to build forecasting models of near-surface wind speed over five locations in northeastern Brazil. Taking advantage of the symbolic nature of the models built, we conducted an extensive series of post-analyses. Overall, our models reduced the forecasting errors by 7%–56% when compared with other techniques, including a real-world operational ensemble model used in Brazil.     

逾期风险预测的宽度和深度学习

逾期风险控制是信用贷款服务的关键业务环节,直接影响放贷企业的收益率和坏账率。随着移动互联网的发展,信贷类金融服务已经惠及普罗大众,逾期风控也从以往依赖规则的人工判断,转为利用大量客户数据构建的信贷模型,以预测客户的逾期概率。相关模型包括传统的机器学习模型和深度学习模型,前者可解释性强、预测能力较弱;后者预测能力强、可解释性较差,且容易发生过拟合。因此,如何融合传统机器学习模型和深度学习模型,一直是信贷数据建模的研究热点。受到推荐系统中宽度和深度学习模型的启发,信贷模型首先可以使用传统机器学习来捕捉结构化数据的特征,同时使用深度学习来捕捉非结构化数据的特征,然后合并两部分学习得到的特征,将其经过线性变换后,最后得到预测的客户的逾期概率。所提模型中和了传统机器学习模型和深度学习模型的优点。实验结果表明,其具有更强的预测客户逾期概率的能力。     

深度强化学习在智能制造中的应用展望综述

深度强化学习作为机器学习发展的最新成果,已经在很多应用领域崭露头角。关于深度强化学习的算法研究和应用研究,产生了很多经典的算法和典型应用领域。深度强化学习应用在智能制造中,能在复杂环境中实现高水平控制。对深度强化学习的研究进行概述,对深度强化学习基本原理进行介绍,包括深度学习和强化学习。介绍深度强化学习算法应用的理论方法,在此基础对深度强化学习的算法进行了分类介绍,分别介绍了基于值函数和基于策略梯度的强化学习算法,列举了这两类算法的主要发展成果,以及其他相关研究成果。对深度强化学习在智能制造的典型应用进行分类分析。对深度强化学习存在的问题和未来发展方向进行了讨论。     

基于深度学习的生理异常检测研究综述

生理信号通常涵盖机体的生物电活动、温度、压力等关键信息，监测其数值波动有助于预警临床事件风险。深度模型是包含多级非线性变换的层级机器学习模型，在特征提取与建模方面优势显著，在计算机辅助诊断领域有着巨大的应用前景。随着连续生理参数监测技术的进步，深度模型在生理电信号异常检测中的效用逐渐提高，研究重点也向临床应用领域拓展。报告了深度模型在生理电信号异常检测中的研究进展。从临床应用出发，分析了经典信号异常检测方法的优势与不足，简述了当前深度模型的建模方式。从判别模型和生成模型的角度总结了经典模型的建模原理及最新应用，同时讨论了深度模型的训练架构和训练策略。结合异常检测在临床中的应用、深度模型的研究进展以及生理数据集的可用性三方面进行总结与讨论，并对未来研究进行展望。     

复杂工业过程特征建模方法及应用研究

针对复杂工业过程的特征建模是研究其优化控制的基础.复杂工业过程普遍具有强干扰、非线性、大时变等诸多不确定性特征,部分工艺涉及复杂的生化反应并伴有强污染和高危性,检测数据具有维度高、噪声大等特性,这均为建立精准的工业模型提出了更急迫的需求和更高的标准.鉴于此,总结并归纳当前复杂工业过程的建模思路和研究进展,旨在从多个视角分析不同建模方法的适用性和有效性,为先进的优化控制理论指导实际工业生产奠定模型基础.首先,从机理建模、数据驱动建模和混合建模3个方向对目前主流的工业建模方法进行划分和综述;然后,阐述各类建模方法的具体设计思路,并分析模型结构和算法特点;接着,调研不同建模策略在解决实际工业过程中的指标建模、被控对象建模、全流程建模等问题的具体应用情况;最后,结合目前工业智能化建设趋势及其面临的挑战性问题,指出未来的研究思路和发展方向.     

A Revisit of Shape Editing Techniques:From the Geometric to the Neural Viewpoint

3D shape editing is widely used in a range of applications such as movie production,computer games and computer aided design.It is also a popular research topic in computer graphics and computer vision.In past decades,researchers have developed a series of editing methods to make the editing process faster,more robust,and more reliable.Traditionally,the deformed shape is determined by the optimal transformation and weights for an energy formulation.With increasing availability of 3D shapes on the Internet,data-driven methods were proposed to improve the editing results.More recently as the deep neural networks became popular,many deep learning based editing methods have been developed in this field,which are naturally data-driven.We mainly survey recent research studies from the geometric viewpoint to those emerging neural deformation techniques and categorize them into organic shape editing methods and man-made model editing methods.Both traditional methods and recent neural network based methods are reviewed.     

Parallel Factories for Smart Industrial Operations: From Big AI Models to Field Foundational Models and Scenarios Engineering

The rapid advancement of fundamental theories and computing capacity has brought artificial intelligence, internet of things, extended reality, and many other new intelligent technologies into our daily lives. Due to the lack of interpretability and reliability guarantees, it is extremely challenging to apply these technologies directly to real-world industrial systems. Here we present a new paradigm for establishing parallel factories in metaverses to accelerate the deployment of intelligent technologies in real-world industrial systems: QAII-1.0. Based on cyber-physical-social systems, QAII-1.0 incorporates complex social and human factors into the design and analysis of industrial operations and is capable of handling industrial operations involving complex social and human behaviors. In QAII-1.0, a field foundational model called EuArtisan combined with scenarios engineering is developed to improve the intelligence of industrial systems while ensuring industrial interpretability and reliability. Finally, parallel oil fields in metaverses are established to demonstrate the operating procedure of QAII-1.0.      

Data-driven soft sensor development based on deep learning technique

In industrial process control, some product qualities and key variables are always difficult to measure online due to technical or economic limitations. As an effective solution, data-driven soft sensors provide stable and reliable online estimation of these variables based on historical measurements of easy-to-measure process variables. Deep learning, as a novel training strategy for deep neural networks, has recently become a popular data-driven approach in the area of machine learning. In the present study, the deep learning technique is employed to build soft sensors and applied to an industrial case to estimate the heavy diesel 95% cut point of a crude distillation unit (CDU). The comparison of modeling results demonstrates that the deep learning technique is especially suitable for soft sensor modeling because of the following advantages over traditional methods. First, with a complex multi-layer structure, the deep neural network is able to contain richer information and yield improved representation ability compared with traditional data-driven models. Second, deep neural networks are established as latent variable models that help to describe highly correlated process variables. Third, the deep learning is semi-supervised so that all available process data can be utilized. Fourth, the deep learning technique is particularly efficient dealing with massive data in practice.     

关于深度学习的综述与讨论

机器学习是通过计算模型和算法从数据中学习规律的一门学问,在各种需要从复杂数据中挖掘规律的领域中有很多应用,已成为当今广义的人工智能领域最核心的技术之一。近年来,多种深度神经网络在大量机器学习问题上取得了令人瞩目的成果,形成了机器学习领域最亮眼的一个新分支——深度学习,也掀起了机器学习理论、方法和应用研究的一个新高潮。对深度学习代表性方法的核心原理和典型优化算法进行了综述,回顾与讨论了深度学习与以往机器学习方法之间的联系与区别,并对深度学习中一些需要进一步研究的问题进行了初步讨论。