检索结果-内蒙古大学图书馆

Reframing control methods for parameters optimization in adversarial image generation

NEURAL NETWORKS 2022年 153卷 303-313页

作者： Alfalouji, Qamar Sartor, Piergiorgio Zanuttigh, Pietro Graz Univ Technol Inffeldgasse 16B A-8010 Graz Austria Sony Europe BV R&D Ctr Europe SL1 Hedelfinger Str 61 D-70327 Stuttgart Germany Univ Padua Dept Informat Engn Via Gradenigo 6B I-35131 Padua Italy

Training procedures for deep networks require the setting of several hyper-parameters that strongly affect the obtained results. The problem is even worse in adversarial learning strategies used for image generation where a proper balancing of the discriminative and generative networks is fundamental for an effective training. In this work we propose a novel hyper-parameters optimization strategy based on the use of Proportional-Integral (PI) and Proportional-Integral-Derivative (PID) controllers. Both open loop and closed loop schemes for the tuning of a single parameter or of multiple parameters together are proposed allowing an efficient parameter tuning without resorting to computationally demanding trial-and-error schemes. We applied the proposed strategies to the widely used BEGAN and CycleGAN models: They allowed to achieve a more stable training that converges faster. The obtained images are also sharper with a slightly better quality both visually and according to the FID and FCN metrics. Image translation results also showed better background preservation and less color artifacts with respect to CycleGAN. (c) 2022 Elsevier Ltd. All rights reserved.

关键词： hyper-parameters optimization GAN Control methods

来源：评论

学校读者我要写书评

暂无评论

Improving the prediction of continuous integration build failures using deep learning

引用

AUTOMATED SOFTWARE ENGINEERING 2022年第1期29卷 21页

作者： Saidani, Islem Ouni, Ali Mkaouer, Mohamed Wiem Univ Quebec ETS Montreal Montreal PQ Canada Rochester Inst Technol Rochester NY 14623 USA

Continuous Integration (CI) aims at supporting developers in integrating code changes constantly and quickly through an automated build process. However, the build process is typically time and resource-consuming as running failed builds can take hours until discovering the breakage;which may cause disruptions in the development process and delays in the product release dates. Hence, preemptively detecting when a software state is most likely to trigger a failure during the build is of crucial importance for developers. Accurate build failures prediction techniques can cut the expenses of CI build cost by early predicting its potential failures. However, developing accurate prediction models is a challenging task as it requires learning long- and short-term dependencies in the historical CI build data as well as extensive feature engineering to derive informative features to learn from. In this paper, we introduce DL-CIBuild a novel approach that uses Long Short-Term Memory (LSTM)-based Recurrent Neural Networks (RNN) to construct prediction models for CI build outcome prediction. The problem is comprised of a single series of CI build outcomes and a model is required to learn from the series of past observations to predict the next CI build outcome in the sequence. In addition, we tailor Genetic Algorithm (GA) to tune the hyper-parameters for our LSTM model. We evaluate our approach and investigate the performance of both cross-project and online prediction scenarios on a benchmark of 91,330 CI builds from 10 large and long-lived software projects that use the Travis CI build system. The statistical analysis of the obtained results shows that the LSTM-based model outperforms traditional Machine Learning (ML) models with both online and cross-project validations. DL-CIBuild has shown also a less sensitivity to the training set size and an effective robustness to the concept drift. Additionally, by considering several hyper-Parameter optimization (HPO) methods as bas

关键词： Continuous integration Build prediction Travis CI Genetic algorithm Long short term memory Machine learning hyper-parameters optimization Concept drift

来源：评论

学校读者我要写书评

暂无评论

Vehicle detection using improved region convolution neural network for accident prevention in smart roads

引用

PATTERN RECOGNITION LETTERS 2022年 158卷 42-47页

作者： Djenouri, Youcef Belhadi, Asma Srivastava, Gautam Djenouri, Djamel Lin, Jerry Chun-Wei SINTEF Digital Math & Cybernet Oslo Norway Kristiania Univ Coll Dept Technol Oslo Norway Brandon Univ Dept Math & Comp Sci Brandon MB Canada Univ West England Comp Sci Res Ctr Dept Comp Sci & Creat Technol Bristol Glos England Western Norway Univ Appl Sci Dept Comp Sci Elect Engn & Math Sci Bergen Norway China Med Univ Res Ctr Interneural Comp Taichung Taiwan

This paper explores the vehicle detection problem and introduces an improved regional convolution neural network. The vehicle data (set of images) is first collected, from which the noise (set of outlier images) is removed using the SIFT extractor. The region convolution neural network is then used to detect the vehicles. We propose a new hyper-parameters optimization model based on evolutionary computation that can be used to tune parameters of the deep learning framework. The proposed solution was tested using the well-known boxy vehicle detection data , which contains more than 20 0,0 0 0 vehicle images and 1,990,0 0 0 annotated vehicles. The results are very promising and show superiority over many current state-of-the-art solutions in terms of runtime and accuracy performances. (c) 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license ( http://***/licenses/by/4.0/ )

关键词： Deep learning Vehicle detection Region convolution neural network hyper-parameters optimization Deep learning Vehicle detection Region convolution neural network hyper-parameters optimization

来源：评论

学校读者我要写书评

暂无评论

An optimized and interpretable carbon price prediction: Explainable deep learning model

引用

CHAOS SOLITONS & FRACTALS 2024年 188卷

作者： Sayed, Gehad Ismail El-Latif, Eman I. Abd Darwish, Ashraf Snasel, Vaclav Hassanien, Aboul Ella Canadian Int Coll Sch Comp Sci Cairo Egypt Benha Univ Fac Sci Banha Egypt Helwan Univ Fac Sci Cairo Egypt VSB Tech Univ Ostrava Fac Elect Engn & Comp Sci Ostrava Czech Republic Kuwait Univ Coll Business Adm POB 5486104 Shidadiya Off B2 Safat 13055 Kuwait Cairo Univ Fac Comp & Artificial Intelligence Cairo Egypt

Accurate prediction of carbon prices is crucial for a stable market, enabling informed decision-making and strategic planning. Over the years, several models for predicting carbon prices have been proposed to enhance accuracy. However, previous research has primarily focused on improving accuracy, often neglecting the importance of making the findings understandable and meaningful. This paper aims to bridge that gap by not only improving prediction accuracy but also ensuring that the results are transparent and comprehensible, thus contributing to more effective and informed decision-making in the carbon market. An optimized Long Short-Term Memory (LSTM) network enhanced with the modified light spectrum optimizer (MLSO) is proposed to improve carbon price prediction accuracy. Additionally, the paper incorporates Explainable AI (XAI) techniques to interpret the results, bridging the gap between accuracy and interpretability. The proposed model is evaluated on carbon price historical transaction data acquired from *** and tested on eight other benchmark datasets with different characteristics. The proposed model achieved 0.66 root mean square error (RMSE), 0.99 R-2, 0.37 mean absolute error (MAE), 0.15 mean absolute percentage error (MAPE), and 0.44 mean square error (MSE). The results showed that low price, high price, and open price features have the highest significance in driving the model's predictions in comparison to other features like date, volume, and price change features. Additionally, the results indicate that the year, day, and month do not significantly influence the carbon price. The proposed model outperforms state-of-the-art models and other well-known machine learning algorithms according to the experimental results. Moreover, the results indicate that the predictive capability of the proposed model serves as a valuable tool for investors and carbon traders to understand the factors influencing price changes, optimize their strategies, and

关键词： Carbon price Carbon emission Light spectrum optimizer Explainable deep-learning Long short term memory hyper-parameters optimization

来源：评论

学校读者我要写书评

暂无评论

NSGA-II as feature selection technique and AdaBoost classifier for COVID-19 prediction using patient's symptoms

引用

NONLINEAR DYNAMICS 2021年第2期106卷 1453-1475页

作者： Soui, Makram Mansouri, Nesrine Alhamad, Raed Kessentini, Marouane Ghedira, Khaled Saudi Elect Univ Coll Comp & Informat Riyadh Saudi Arabia Univ Gabes Gabes Tunisia Univ Michigan Dearborn MI 48128 USA Private Higher Sch Engn & Technol Ariana Tunisia

Nowadays, humanity is facing one of the most dangerous pandemics known as COVID-19. Due to its high inter-person contagiousness, COVID-19 is rapidly spreading across the world. Positive patients are often suffering from different symptoms that can vary from mild to severe including cough, fever, sore throat, and body aches. In more dire cases, infected patients can experience severe symptoms that can cause breathing difficulties which lead to stern organ failure and die. The medical corps all over the world are overloaded because of the exponentially myriad number of contagions. Therefore, screening for the disease becomes overwrought with the limited tools of test. Additionally, test results may take a long time to acquire, leaving behind a higher potential for the prevalence of the virus among other individuals by the patients. To reduce the chances of infection, we suggest a prediction model that distinguishes the infected COVID-19 cases based on clinical symptoms and features. This model can be helpful for citizens to catch their infection without the need for visiting the hospital. Also, it helps the medical staff in triaging patients in case of a deficiency of medical amenities. In this paper, we use the non-dominated sorting genetic algorithm (NSGA-II) to select the interesting features by finding the best trade-offs between two conflicting objectives: minimizing the number of features and maximizing the weights of selected features. Then, a classification phase is conducted using an AdaBoost classifier. The proposed model is evaluated using two different datasets. To maximize results, we performed a natural selection of hyper-parameters of the classifier using the genetic algorithm. The obtained results prove the efficiency of NSGA-II as a feature selection algorithm combined with AdaBoost classifier. It exhibits higher classification results that outperformed the existing methods.

关键词： COVID-19 prediction Machine learning Feature selection AdaBoost NSGA-II hyper-parameters optimization

来源：评论

学校读者我要写书评

暂无评论

Application of global optimization methods to model and feature selection

引用

PATTERN RECOGNITION 2012年第10期45卷 3676-3686页

作者： Boubezoul, Abderrahmane Paris, Sebastien Aix Marseille Univ Lab Sci Informat & Syst LSIS DYNI UMR CNRS 7296 F-13397 Marseille 20 France Paris Est Univ IFSTTAR IM LEPSIS F-75732 Paris France

Many data mining applications involve the task of building a model for predictive classification. The goal of this model is to classify data instances into classes or categories of the same type. The use of variables not related to the classes can reduce the accuracy and reliability of classification or prediction model. Superfluous variables can also increase the costs of building a model particularly on large datasets. The feature selection and hyper-parameters optimization problem can be solved by either an exhaustive search over all parameter values or an optimization procedure that explores only a finite subset of the possible values. The objective of this research is to simultaneously optimize the hyper-parameters and feature subset without degrading the generalization performances of the induction algorithm. We present a global optimization approach based on the use of Cross-Entropy Method to solve this kind of problem. (C) 2012 Elsevier Ltd. All rights reserved.

关键词： Cross-Entropy Method Feature selection Particle swarm optimization hyper-parameters optimization Support vector machines

来源：评论

学校读者我要写书评

暂无评论

A genetic Gaussian process regression model based on memetic algorithm

引用

Journal of Central South University 2013年第11期20卷 3085-3093页

作者：张乐刘忠张建强任雄伟 College of Electronic Naval University of Engineering Wuhan Mechanical Technology College

Gaussian process(GP)has fewer parameters,simple model and output of probabilistic sense,when compared with the methods such as support vector *** of the hyper-parameters is critical to the performance of Gaussian process ***,the common-used algorithm has the disadvantages of difficult determination of iteration steps,over-dependence of optimization effect on initial values,and easily falling into local *** solve this problem,a method combining the Gaussian process with memetic algorithm was *** on this method,memetic algorithm was used to search the optimal hyper parameters of Gaussian process regression(GPR)model in the training process and form MA-GPR algorithms,and then the model was used to predict and test the *** used in the marine long-range precision strike system(LPSS)battle effectiveness evaluation,the proposed MA-GPR model significantly improved the prediction accuracy,compared with the conjugate gradient method and the genetic algorithm optimization process.

关键词： Gaussian process hyper-parameters optimization memetic algorithm regression model

来源：评论

学校读者我要写书评

暂无评论

Early diagnosis of Parkinson's disease from multiple voice recordings by simultaneous sample and feature selection

引用

EXPERT SYSTEMS WITH APPLICATIONS 2019年 137卷 22-28页

作者： Ali, Liaqat Zhu, Ce Zhou, Mingyi Liu, Yipeng UESTC Sch Informat & Commun Engn Chengdu Sichuan Peoples R China Univ Sci & Technol Dept Elect Engn Bannu Pakistan

Parkinson's disease (PD) is a serious neurodegenerative disorder. It is reported that more than 90% of PD patients have voice impairments. Multiple types of voice recordings have been used for PD detection. Previous work indicates that the use of multiple types of samples per subject degenerates PD detection accuracy. In this paper, we validate it, and propose a two dimensional data selection method for sample and feature selection. The proposed method ranks features by using chi-square statistical model, searches optimal subset of the ranked features and iteratively selects samples. Experimental results show that the proposed method outperforms the state-of-the-art methods in terms of PD detection accuracy on multiple types of voice data. (C) 2019 Elsevier Ltd. All rights reserved.

关键词： Deep neural network Feature selection hyper-parameters optimization Parkinson'S disease Sample selection

来源：评论

学校读者我要写书评

暂无评论

Real-Time Implementation of Quadrotor UAV Control System Based on a Deep Reinforcement Learning Approach

引用

Computers, Materials & Continua 2024年第12期81卷 4757-4786页

作者： Taha Yacine Trad Kheireddine Choutri Mohand Lagha Souham Meshoul Fouad Khenfri Raouf Fareh Hadil Shaiba Aeronautical Sciences Laboratory Aeronautical and Spatial Studies InstituteBlida 1 UniversityBlida0900Algeria Department of Information Technology College of Computer and Information SciencesPrincess Nourah Bint Abdulrahman UniversityRiyadh11671Saudi Arabia Energy and Embedded Systems for Transportation Research Department ESTACA-LABMontigny-Le-Bretonneux78066France Department of Electrical Engineering University of SharjahSharjah27272United Arab Emirates Department of Computer Science College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityRiyadh11671Saudi Arabia

The popularity of quadrotor Unmanned Aerial Vehicles(UAVs)stems from their simple propulsion systems and structural ***,their complex and nonlinear dynamic behavior presents a significant challenge for control,necessitating sophisticated algorithms to ensure stability and accuracy in *** strategies have been explored by researchers and control engineers,with learning-based methods like reinforcement learning,deep learning,and neural networks showing promise in enhancing the robustness and adaptability of quadrotor control *** paper investigates a Reinforcement Learning(RL)approach for both high and low-level quadrotor control systems,focusing on attitude stabilization and position tracking tasks.A novel reward function and actor-critic network structures are designed to stimulate high-order observable states,improving the agent’s understanding of the quadrotor’s dynamics and environmental *** address the challenge of RL hyper-parameter tuning,a new framework is introduced that combines Simulated Annealing(SA)with a reinforcement learning algorithm,specifically Simulated Annealing-Twin Delayed Deep Deterministic Policy Gradient(SA-TD3).This approach is evaluated for path-following and stabilization tasks through comparative assessments with two commonly used control methods:Backstepping and Sliding Mode Control(SMC).While the implementation of the well-trained agents exhibited unexpected behavior during real-world testing,a reduced neural network used for altitude control was successfully implemented on a Parrot Mambo mini *** results showcase the potential of the proposed SA-TD3 framework for real-world applications,demonstrating improved stability and precision across various test scenarios and highlighting its feasibility for practical deployment.

关键词： Deep reinforcement learning hyper-parameters optimization path following quadrotor twin delayed deep deterministic policy gradient and simulated annealing

来源：评论

学校读者我要写书评

暂无评论

Application of Meta-Heuristic Algorithms for Training Neural Networks and Deep Learning Architectures: A Comprehensive Review

引用

NEURAL PROCESSING LETTERS 2023年第4期55卷 4519-4622页

作者： Kaveh, Mehrdad Mesgari, Mohammad Saadi KN Toosi Univ Technol Dept Geodesy & Geomat Tehran *** Iran

The learning process and hyper-parameter optimization of artificial neural networks (ANNs) and deep learning (DL) architectures is considered one of the most challenging machine learning problems. Several past studies have used gradient-based back propagation methods to train DL architectures. However, gradient-based methods have major drawbacks such as stucking at local minimums in multi-objective cost functions, expensive execution time due to calculating gradient information with thousands of iterations and needing the cost functions to be continuous. Since training the ANNs and DLs is an NP-hard optimization problem, their structure and parameters optimization using the meta-heuristic (MH) algorithms has been considerably raised. MH algorithms can accurately formulate the optimal estimation of DL components (such as hyper-parameter, weights, number of layers, number of neurons, learning rate, etc.). This paper provides a comprehensive review of the optimization of ANNs and DLs using MH algorithms. In this paper, we have reviewed the latest developments in the use of MH algorithms in the DL and ANN methods, presented their disadvantages and advantages, and pointed out some research directions to fill the gaps between MHs and DL methods. Moreover, it has been explained that the evolutionary hybrid architecture still has limited applicability in the literature. Also, this paper classifies the latest MH algorithms in the literature to demonstrate their effectiveness in DL and ANN training for various applications. Most researchers tend to extend novel hybrid algorithms by combining MHs to optimize the hyper-parameters of DLs and ANNs. The development of hybrid MHs helps improving algorithms performance and capable of solving complex optimization problems. In general, the optimal performance of the MHs should be able to achieve a suitable trade-off between exploration and exploitation features. Hence, this paper tries to summarize various MH algorithms in terms of th

关键词： Deep learning (DL) Artificial neural networks (ANN) Meta-heuristics (MH) hyper-parameters optimization Training And gradient-based back propagation (BP) learning algorithm

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：