Intelligent Feature Selection with Deep Learning Based Financial Risk Assessment Model

Due to global financial crisis, risk management has received significant attention to avoid loss and maximize profit in any business. Since the financial crisis prediction (FCP) process is mainly based on data driven decision making and intelligent models, artificial intelligence (AI) and machine learning (ML) models are widely utilized. This article introduces an intelligent feature selection with deep learning based financial risk assessment model (IFSDL-FRA). The proposed IFSDL-FRA technique aims to determine the financial crisis of a company or enterprise. In addition, the IFSDL-FRA technique involves the design of new water strider optimization algorithm based feature selection (WSOA-FS) manner to an optimum selection of feature subsets. Moreover, Deep Random Vector Functional Link network (DRVFLN) classification technique was applied to properly allot the class labels to the financial data. Furthermore, improved fruit fly optimization algorithm (IFFOA) based hyperparameter tuning process is carried out to optimally tune the hyperparameters of the DRVFLN model. For enhancing the better performance of the IFSDL-FRA technique, an extensive set of simulations are implemented on benchmark financial datasets and the obtained outcomes determine the betterment of IFSDL-FRA technique on the recent state of art approaches.     

Optimizing Steering Angle Predictive Convolutional Neural Network for Autonomous Car

Deep learning techniques, particularly convolutional neural networks (CNNs), have exhibited remarkable performance in solving vision-related problems, especially in unpredictable, dynamic, and challenging environments. In autonomous vehicles, imitation-learning-based steering angle prediction is viable due to the visual imagery comprehension of CNNs. In this regard, globally, researchers are currently focusing on the architectural design and optimization of the hyperparameters of CNNs to achieve the best results. Literature has proven the superiority of metaheuristic algorithms over the manual-tuning of CNNs. However, to the best of our knowledge, these techniques are yet to be applied to address the problem of imitation-learning-based steering angle prediction. Thus, in this study, we examine the application of the bat algorithm and particle swarm optimization algorithm for the optimization of the CNN model and its hyperparameters, which are employed to solve the steering angle prediction problem. To validate the performance of each hyperparameters’ set and architectural parameters’ set, we utilized the Udacity steering angle dataset and obtained the best results at the following hyperparameter set: optimizer, Adagrad; learning rate, 0.0052; and nonlinear activation function, exponential linear unit. As per our findings, we determined that the deep learning models show better results but require more training epochs and time as compared to shallower ones. Results show the superiority of our approach in optimizing CNNs through metaheuristic algorithms as compared with the manual-tuning approach. Infield testing was also performed using the model trained with the optimal architecture, which we developed using our approach.     

基于SSPSO优化GRNN的水电站厂房结构振动响应预测

提出基于优胜劣汰、步步选择的粒子群优化算法(SSPSO),弥补了一般粒子群优化算法容易陷入局部极值、早熟收敛或停滞的缺陷。并运用SSPSO对广义回归神经网络(GRNN)平滑参数P进行优化,充分利用SSPSO寻优能力强及径向基函数调整参数少的优点,建立厂房结构的振动响应预测模型,对某厂顶溢流式水电站的厂坝结构振动响应问题展开预测研究。通过分析预测效果得出:与一般的粒子群算法相比,所提出的SSPSO算法的寻优能力得到了很大的提高。与此同时,基于SSPSO优化的广义回归神经网络(SSPSO-GRNN)与其他网络相比,在预测精度、收敛性能、泛化能力等各个方面得到了很大提升。为水电站厂房振动响应预测提供了新的方法和思路,为增强厂房结构的智能化监测提供了保障。     

IoT-Cloud Assisted Botnet Detection Using Rat Swarm Optimizer withDeepLearning

Nowadays, Internet of Things (IoT) has penetrated all facets of human life while on the other hand, IoT devices are heavily prone to cyberattacks. It has become important to develop an accurate system that can detect malicious attacks on IoT environments in order to mitigate security risks. Botnet is one of the dreadful malicious entities that has affected many users for the past few decades. It is challenging to recognize Botnet since it has excellent carrying and hidden capacities. Various approaches have been employed to identify the source of Botnet at earlier stages. Machine Learning (ML) and Deep Learning (DL) techniques are developed based on heavy influence from Botnet detection methodology. In spite of this, it is still a challenging task to detect Botnet at early stages due to low number of features accessible from Botnet dataset. The current study devises IoT with Cloud Assisted Botnet Detection and Classification utilizing Rat Swarm Optimizer with Deep Learning (BDC-RSODL) model. The presented BDC-RSODL model includes a series of processes like pre-processing, feature subset selection, classification, and parameter tuning. Initially, the network data is pre-processed to make it compatible for further processing. Besides, RSO algorithm is exploited for effective selection of subset of features. Additionally, Long Short Term Memory (LSTM) algorithm is utilized for both identification and classification of botnets. Finally, Sine Cosine Algorithm (SCA) is executed for fine-tuning the hyperparameters related to LSTM model. In order to validate the promising performance of BDC-RSODL system, a comprehensive comparison analysis was conducted. The obtained results confirmed the supremacy of BDC-RSODL model over recent approaches.     

Hybridizing the harmony search algorithm with a spreadsheet ‘Solver’ for solving continuous engineering optimization problems

In this article, a hybrid global–local optimization algorithm is proposed to solve continuous engineering optimization problems. In the proposed algorithm, the harmony search (HS) algorithm is used as a global-search method and hybridized with a spreadsheet ‘Solver’ to improve the results of the HS algorithm. With this purpose, the hybrid HS–Solver algorithm has been proposed. In order to test the performance of the proposed hybrid HS–Solver algorithm, several unconstrained, constrained, and structural-engineering optimization problems have been solved and their results are compared with other deterministic and stochastic solution methods. Also, an empirical study has been carried out to test the performance of the proposed hybrid HS–Solver algorithm for different sets of HS solution parameters. Identified results showed that the hybrid HS–Solver algorithm requires fewer iterations and gives more effective results than other deterministic and stochastic solution algorithms.     

基于贝叶斯优化与改进LeNet-5的滚动轴承故障诊断

考虑到卷积神经网络在滚动轴承故障诊断中存在网络结构难以确定、训练次数过多、时间过长等问题,设计了一种贝叶斯优化改进LeNet-5算法,以及采用该算法构建的轴承故障诊断模型。采用贝叶斯优化训练过程中学习率等超参数,多种故障轴承的振动信号直接作为改进LeNet-5网络的输入,对池化输出采用批归一化处理和改进池化层激活函数防止过拟合,利用全局平均池化层替代全连接层提高改进LeNet-5网络的泛化能力,用Softmax分类器实现滚动轴承故障的分类。通过轴承数据库开展实验,实验表明,该算法构建的轴承故障诊断模型在训练集上准确率为99.94%,验证集上的准确率为99.89%,测试集准确率也达到99.65%,与一维卷积神经网络和二维卷积神经网络对比分析,基于贝叶斯优化改进LeNet-5算法构建的轴承故障诊断模型在滚动轴承的故障诊断模型具有更高的准确率,更少的训练次数和训练时间。     

MDTS: automatic complex materials design using Monte Carlo tree search

Abstract

Complex materials design is often represented as a black-box combinatorial optimization problem. In this paper, we present a novel python library called MDTS (Materials Design using Tree Search). Our algorithm employs a Monte Carlo tree search approach, which has shown exceptional performance in computer Go game. Unlike evolutionary algorithms that require user intervention to set parameters appropriately, MDTS has no tuning parameters and works autonomously in various problems. In comparison to a Bayesian optimization package, our algorithm showed competitive search efficiency and superior scalability. We succeeded in designing large Silicon-Germanium (Si-Ge) alloy structures that Bayesian optimization could not deal with due to excessive computational cost. MDTS is available at https://github.com/tsudalab/MDTS.     

Data Mining with Comprehensive Oppositional Based Learning for Rainfall Prediction

Data mining process involves a number of steps from data collection to visualization to identify useful data from massive data set. the same time, the recent advances of machine learning (ML) and deep learning (DL) models can be utilized for effectual rainfall prediction. With this motivation, this article develops a novel comprehensive oppositional moth flame optimization with deep learning for rainfall prediction (COMFO-DLRP) Technique. The proposed CMFO-DLRP model mainly intends to predict the rainfall and thereby determine the environmental changes. Primarily, data pre-processing and correlation matrix (CM) based feature selection processes are carried out. In addition, deep belief network (DBN) model is applied for the effective prediction of rainfall data. Moreover, COMFO algorithm was derived by integrating the concepts of comprehensive oppositional based learning (COBL) with traditional MFO algorithm. Finally, the COMFO algorithm is employed for the optimal hyperparameter selection of the DBN model. For demonstrating the improved outcomes of the COMFO-DLRP approach, a sequence of simulations were carried out and the outcomes are assessed under distinct measures. The simulation outcome highlighted the enhanced outcomes of the COMFO-DLRP method on the other techniques.     

Logistic 组稀疏回归模型的 Bayes 建模及变分推断

在工程应用中,如数据挖掘、成本预测以及风险预测等,Logistic 回归是一类十分重要的预测方法．当前,大部分 Logistic 回归方法都是基于优化准则而设计,这类回归方法具有参数调试过程繁琐、模型解释性差、估计子没有置信区间等缺点．本文从 Bayes 概率角度研究 Logistic 组稀疏性回归的建模与推断问题．具体来说,首先利用高斯-方差混合公式提出 Logistic 组稀疏回归的 Bayes 概率模型;其次,通过变分 Bayes 方法设计出一个高效的推断算法．在模拟数据上的实验结果表明,本文所提出的方法具有较好的预测性能．     

Target Detection Algorithm in Crime Recognition Using Artificial Intelligence

Presently, suspect prediction of crime scenes can be considered as a classification task, which predicts the suspects based on the time, space, and type of crime. Performing digital forensic investigation in a big data environment poses several challenges to the investigational officer. Besides, the facial sketches are widely employed by the law enforcement agencies for assisting the suspect identification of suspects involved in crime scenes. The sketches utilized in the forensic investigations are either drawn by forensic artists or generated through the computer program (composite sketches) based on the verbal explanation given by the eyewitness or victim. Since this suspect identification process is slow and difficult, it is required to design a technique for a quick and automated facial sketch generation. Machine Learning (ML) and deep learning (DL) models find it useful to automatically support the decision of forensics experts. The challenge is the incorporation of the domain expert knowledge with DL models for developing efficient techniques to make better decisions. In this view, this study develops a new artificial intelligence (AI) based DL model with face sketch synthesis (FSS) for suspect identification (DLFSS-SI) in a big data environment. The proposed method performs preprocessing at the primary stage to improvise the image quality. In addition, the proposed model uses a DL based MobileNet (MN) model for feature extractor, and the hyper parameters of the MobileNet are tuned by quasi oppositional firefly optimization (QOFFO) algorithm. The proposed model automatically draws the sketches of the input facial images. Moreover, a qualitative similarity assessment takes place with the sketch drawn by a professional artist by the eyewitness. If there is a higher resemblance between the two sketches, the suspect will be determined. To validate the effective performance of the DLFSS-SI method, a detailed qualitative and quantitative examination takes place. The experimental outcome stated that the DLFSS-SI model has outperformed the compared methods in terms of mean square error (MSE), peak signal to noise ratio (PSNR), average actuary, and average computation time.     

The Role of Deep Learning in Parking Space Identification andPredictionSystems

In today’s smart city transportation, traffic congestion is a vexing issue, and vehicles seeking parking spaces have been identified as one of the causes leading to approximately 40% of traffic congestion. Identifying parking spaces alone is insufficient because an identified available parking space may have been taken by another vehicle when it arrives, resulting in the driver’s frustration and aggravating traffic jams while searching for another parking space. This explains the need to predict the availability of parking spaces. Recently, deep learning (DL) has been shown to facilitate drivers to find parking spaces efficiently, leading to a promising performance enhancement in parking identification and prediction systems. However, no work reviews DL approaches applied to solve parking identification and prediction problems. Inspired by this gap, the purpose of this work is to investigate, highlight, and report on recent advances in DL approaches applied to predict and identify the availability of parking spaces. A taxonomy of DL-based parking identification and prediction systems is established as a methodology by classifying and categorizing existing literature, and by doing so, the salient and supportive features of different DL techniques for providing parking solutions are presented. Moreover, several open research challenges are outlined. This work identifies that there are various DL architectures, datasets, and performance measures used to address parking identification and prediction problems. Moreover, there are some open-source implementations available that can be used directly either to extend existing works or explore a new domain. This is the first short survey article that focuses on the use of DL-based techniques in parking identification and prediction systems for smart cities. This study concludes that although the deployment of DL in parking identification and prediction systems provides various benefits, the convergence of these two types of systems and DL brings about new issues that must be resolved in the near future.     

Introducing two hyperparameters in Bayesian estimation of wave spectra

An estimate of the on-site wave spectrum can be obtained from measured ship responses by use of Bayesian modelling, which means that the wave spectrum is found as the optimum solution from a probabilistic viewpoint. The paper describes the introduction of two hyperparameters into Bayesian modelling so that the prior information included in the modelling is based on two constraints: the wave spectrum must be smooth directional-wise as well as frequency-wise. Traditionally, only one hyperparameter has been used to control the amount of smoothing applied in both the frequency and directional ranges. From numerical simulations of stochastic response measurements, it is shown that the optimal hyperparameters, determined by use of ABIC (a Bayesian Information Criterion), correspond to the best estimate of the wave spectrum, which is not always the case when only one hyperparameter is included in the Bayesian modelling. The paper includes also an analysis of full-scale motion measurements where wave spectra estimated by the Bayesian modelling are compared with results from ocean surface measurements by satellite and from a wave radar. The agreement is found to be reasonable.     

A Bayesian Approach to Sequential Optimization based on Computer Experiments

Computer experiments are used frequently for the study and improvement of a process under study. Optimizing such process based on a computer model is costly, so an approximation of the computer model, or metamodel, is used. Efficient global optimization (EGO) is a sequential optimization method for computer experiments based on a Gaussian process model approximation to the computer model response. A long‐standing problem in EGO is that it does not consider the uncertainty in the parameter estimates of the Gaussian process. Treating these estimates as if they are the true parameters leads to an improper assessment of the precision of the approximation, a precision that is crucial to assess not only in optimization but in metamodeling in general. One way to account for these uncertainties is to use bootstrapping, studied by previous authors. Alternatively, some other authors have mentioned how a Bayesian approach may be the best way to incorporate the parameter uncertainty in the optimization, but no fully Bayesian approach for EGO has been implemented in practice. In this paper, we present a fully Bayesian implementation of the EGO method. The proposed Bayesian EGO algorithm is validated through simulation of noisy nonlinear functions and compared with the standard EGO method and the bootstrapped EGO. We also apply the Bayesian EGO algorithm to the optimization of a stochastic computer model. It is shown how a Bayesian approach to EGO allows one to optimize any function of the posterior predictive density. Copyright © 2014 John Wiley & Sons, Ltd.     

A study on the application of material selection optimization approach for structural-acoustic optimization

Issues of application of the material selection optimization approach for structural-acoustic optimization is investigated herein. By introducing the stacking sequence hypothesis of metal material, the mechanical properties parameters and plies’ numbers of the metal material or composite material are defined as design variables; the mathematical formulation about material selection optimization approach is established. Finally, a hexahedral box structure is taken as an example, and the material selection optimization is conducted. By introducing genetic algorithm (GA), the optimization problem is solved. The numerical example shows the effectiveness of the proposed stacking sequence hypothesis of metal material.     

Fast antenna subset selection algorithms for multiple-input multiple-output relay systems

The antenna subset selection technique balances the performance and hardware cost in the multipleinput multiple-output (MIMO) systems, and the problems on the antenna selection in MIMO relay systems have not been fully solved. This paper considers antenna selection on amplify-and-forward (AF) and decode-andforward (DF) MIMO relay systems to maximise capacity. Since the optimal antenna selection algorithm has high complexity, two fast algorithms are proposed. The selection criterion of the algorithm for AF relay is to maximise a lower bound of the capacity, but not the exact capacity. This criterion reduces algorithmic complexity. The algorithm for DF relay is an extension of an existing antenna subset selection algorithm for one-hop MIMO systems. The authors show the derivations of the algorithms in detail, and analyse their complexity in terms of numbers of complex multiplications. Simulation results show that the proposed algorithms for both cases achieve comparable performance to the optimal algorithm under various conditions, and have decreased complexity.     

Optimal design of multi-state weighted k-out-of-n systems based on component design

This paper presents a study on design optimization of multi-state weighted k-out-of-n systems. The studied system reliability model is more general than the traditional k-out-of-n system model. The system and its components are capable of assuming a whole range of performance levels, varying from perfect functioning to complete failure. A utility value corresponding to each state is used to indicate the corresponding performance level. A widely studied reliability optimization problem is the “component selection problem”, which involves selection of components with known reliability and cost characteristics. Less adequately addressed has been the problem of determining system cost and utility based on the relationships between component reliability, cost and utility. This paper addresses this topic. All the optimization problems dealt with in this paper can be categorized as either minimizing the expected total system cost subject to system reliability requirements, or maximizing system reliability subject to total system cost limitation. The resulting optimization problems are too complicated to be solved by traditional optimization approaches; therefore, genetic algorithm (GA) is used to solve them. Our results show that GA is a powerful tool for solving these kinds of problems.     

Ensembles of Deep Learning Framework for Stomach Abnormalities Classification

Abnormalities of the gastrointestinal tract are widespread worldwide today. Generally, an effective way to diagnose these life-threatening diseases is based on endoscopy, which comprises a vast number of images. However, the main challenge in this area is that the process is time-consuming and fatiguing for a gastroenterologist to examine every image in the set. Thus, this led to the rise of studies on designing AI-based systems to assist physicians in the diagnosis. In several medical imaging tasks, deep learning methods, especially convolutional neural networks (CNNs), have contributed to the state-of-the-art outcomes, where the complicated nonlinear relation between target classes and data can be learned and not limit to hand-crafted features. On the other hand, hyperparameters are commonly set manually, which may take a long time and leave the risk of non-optimal hyperparameters for classification. An effective tool for tuning optimal hyperparameters of deep CNN is Bayesian optimization. However, due to the complexity of the CNN, the network can be regarded as a black-box model where the information stored within it is hard to interpret. Hence, Explainable Artificial Intelligence (XAI) techniques are applied to overcome this issue by interpreting the decisions of the CNNs in such wise the physicians can trust. To play an essential role in real-time medical diagnosis, CNN-based models need to be accurate and interpretable, while the uncertainty must be handled. Therefore, a novel method comprising of three phases is proposed to classify these life-threatening diseases. At first, hyperparameter tuning is performed using Bayesian optimization for two state-of-the-art deep CNNs, and then Darknet53 and InceptionV3 features are extracted from these fine-tunned models. Secondly, XAI techniques are used to interpret which part of the images CNN takes for feature extraction. At last, the features are fused, and uncertainties are handled by selecting entropy-based features. The experimental results show that the proposed method outperforms existing methods by achieving an accuracy of 97% based on a Bayesian optimized Support Vector Machine classifier.

     

三水源新安江模型参数不确定性分析PAM算法

针对水文模型参数不确定性分析常用方法 收敛速度缓慢,容易陷入参数空间局部最优区域等 问题,提出了PAM (parallel adaptive metropolis) 算法;对三水源新安江模型参数不确定性进行分析 研究。实例研究表明显著提高了计算速度和求解质 量,参数后验分布结果为区间预报提供了条件。     

A framework for self-evolving computational material models inspired by deep learning

There exists a deep chasm between machine learning (ML) and high-fidelity computational material models in science and engineering. Due to the complex interaction of internal physics, ML methods hardly conquer or innovate them. To fill the chasm, this paper finds an answer from the central notions of deep learning (DL) and proposes information index and link functions, which are essential to infuse principles of physics into ML. Like the convolution process of DL, the proposed information index integrates adjacent information and quantifies the physical similarity between laboratory and reality, enabling ML to see through a complex target system with the perspective of scientists. Like the hidden layers' weights of DL, the proposed link functions unravel the hidden relations between information index and physics rules. Like the error backpropagation of DL, the proposed framework adopts fitness-based spawning scheme of evolutionary algorithm. The proposed framework demonstrates that a fusion of information index, link functions, evolutionary algorithm, and Bayesian update scheme can engender self-evolving computational material models and that the fusion will help rename ML as a partner of researchers in the broad science and engineering.     

Hybrid approach for remaining useful life prediction of ball bearings

Remaining useful life (RUL) prediction plays an important role in predictive maintenance systems to support decision‐makers for arranging maintenance tasks and related resources. We propose a hybrid approach that is combined an exponential weighted moving average (EWMA) control chart for anomaly detection and machine learning models such as support vector regression (SVR) and random forest regression (RFR) with differential evolution (DE) algorithm to predict the RULs of ball bearings. Here, DE algorithm is used to find the optimal hyperparameters of SVR model. The datasets of ball bearings from the Prognostics Data Repository of NASA are used to compare the prediction performance of different methods. The degradation behavior of training data from the anomaly time to the end of life is used to transfer learning for the testing data in the SVR and RFR models. The results indicate that the proposed methods outperform the other four existing methods in terms of score. Therefore, the proposed hybrid approach is a reliable tool for the RUL prediction of ball bearings.