检索结果-南通市图书馆

Performance Evaluation ofmulti-Agent Reinforcement Learning Algorithms

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Intelligent Automation & Soft Computing 2024年第2期39卷 337-352页

作者： abdulghani m.abdulghani mokhles m.abdulghani Wilbur L.Walters Khalid H.Abed Department of Electrical&Computer Engineering and Computer Science Jackson State UniversityJacksonMS 39217USA

multi-Agent Reinforcement Learning(mARL)has proven to be successful in cooperative *** is used to investigate how autonomous agents with the same interests can connect and act in one *** cooperation scenarios are explored in recreational cooperative augmented reality environments,as well as realworld scenarios in *** this paper,we explore the realm of mARL and its potential applications in cooperative *** focus is on developing a multi-agent system that can collaborate to attack or defend against enemies and achieve victory withminimal *** accomplish this,we utilize the StarCraftmulti-Agent Challenge(SmAC)environment and train four mARL algorithms:Q-learning with mixtures of Experts(QmIX),Value-DecompositionNetwork(VDN),multi-agent Proximal PolicyOptimizer(mAPPO),andmulti-Agent Actor Attention Critic(mAA2C).These algorithms allow multiple agents to cooperate in a specific scenario to achieve the targeted *** results show that the QmIX algorithm outperforms the other three algorithms in the attacking scenario,while the VDN algorithm achieves the best results in the defending ***,the VDNalgorithmreaches the highest value of battle wonmean and the lowest value of dead *** research demonstrates the potential formARL algorithms to be used in real-world applications,such as controllingmultiple robots to provide helpful services or coordinating teams of agents to accomplish tasks that would be impossible for a human to *** SmAC environment provides a unique opportunity to test and evaluate mARL algorithms in a challenging and dynamic environment,and our results show that these algorithms can be used to achieve victory with minimal damage.

关键词： Reinforcement learning RL multi-agent mARL SmAC VDN QmIX mAPPO

multiple Data Augmentation Strategy for Enhancing the Performance of YOLOv7 Object Detection Algorithm

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal on Artificial Intelligence 2023年第1期5卷 15-30页

作者： abdulghani m.abdulghani mokhles m.abdulghani Wilbur L.Walters Khalid H.Abed Department of Electrical&Computer Engineering and Computer Science Jackson State UniversityJackson39217USA

The object detection technique depends on various methods for duplicating the dataset without adding more *** augmentation is a popularmethod that assists deep neural networks in achieving better generalization performance and can be seen as a type of implicit *** is recommended in the casewhere the amount of high-quality data is limited,and gaining new examples is costly and *** this paper,we trained YOLOv7 with a dataset that is part of the Open Images dataset that has 8,600 images with four classes(Car,Bus,motorcycle,and Person).We used five different data augmentations techniques for duplicates and improvement of our *** performance of the object detection algorithm was compared when using the proposed augmented dataset with a combination of two and three types of data augmentation with the result of the original *** evaluation result for the augmented data gives a promising result for every object,and every kind of data augmentation gives a different *** mAP@.5 of all classes was 76%,and F1-score was 74%.The proposed method increased the mAP@.5 value by+13%and F1-score by+10%for all objects.

关键词： Artificial intelligence object detection YOLOv7 data augmentation data brightness data darkness data blur data noise convolutional neural network

AI Safety Approach for minimizing Collisions in Autonomous Navigation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal on Artificial Intelligence 2023年第1期5卷 1-14页

作者： abdulghani m.abdulghani mokhles m.abdulghani Wilbur L.Walters Khalid H.Abed Department of Electrical&Computer Engineering and Computer Science Jackson State UniversityJackson39217USA

Autonomous agents can explore the environment around them when equipped with advanced hardware and software systems that help intelligent agents minimize *** systems are developed under the term Artificial Intelligence(AI)*** safety is essential to provide reliable service to consumers in various fields such asmilitary,education,healthcare,and *** paper presents the design of an AI safety algorithmfor safe autonomous navigation using Reinforcement Learning(RL).machine Learning Agents Toolkit(mL-Agents)was used to train the agentwith a proximal policy optimizer algorithmwith an intrinsic curiositymodule(PPO+ICm).This training aims to improve AI safety and minimize or prevent any mistakes that can cause dangerous collisions by the intelligent *** experiments have been executed to validate the results of our *** designed algorithmwas tested in a virtual environment with four differentmodels.A comparison was presented in four cases to identify the best-performing model for improvingAI *** designed algorithmenabled the intelligent agent to perform the required task safely using RL.A goal collision ratio of 64%was achieved,and the collision incidents were minimized from 134 to 52 in the virtual environment within 30min.

关键词： Artificial intelligence AI safety autonomous robots unmanned systems Unity simulations reinforcement learning RL machine learning mL-Agents human-machine teaming

Implementation of Strangely Behaving Intelligent Agents to Determine Human Intervention During Reinforcement Learning

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal on Artificial Intelligence 2022年第4期4卷 261-277页

作者： Christopher C.Rosser Wilbur L.Walters abdulghani m.abdulghani mokhles m.abdulghani Khalid H.Abed Department of Electrical&Computer Engineering and Computer Science Jackson State University(JSU)Jackson39217MSUSA

Intrinsic motivation helps autonomous exploring agents traverse a larger portion of their ***,simulations of different learning environments in previous research show that after millions of timesteps of successful training,an intrinsically motivated agent may learn to act in ways unintended by the *** potential for unintended actions of autonomous exploring agents poses threats to the environment and humans if operated in the real *** investigated this topic by using Unity’s machineLearningAgent Toolkit(mL-Agents)implementation of the Proximal Policy Optimization(PPO)algorithm with the Intrinsic Curiosity module(ICm)to train autonomous exploring agents in three learning *** demonstrate that ICm,although designed to assist agent navigation in environments with sparse reward generation,increasing gradually as a tool for purposely training misbehaving agent in significantly less than 1 million *** present the following achievements:1)experiments designed to cause agents to act undesirably,2)a metric for gauging how well an agent achieves its goal without collisions,and 3)validation of PPO best ***,we used optimized methods to improve the agent’s performance and reduce collisions within the same *** achievements help further our understanding of the significance of monitoring training statistics during reinforcement learning for determining how humans can intervene to improve agent safety and performance.

关键词： Artificial intelligence AI safety reinforcement learning human-inthe-loop intrinsic motivation unity simulations human-machine teaming

New Soliton Wave Solutions to a Nonlinear Equation Arising in Plasma Physics

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computer modeling in Engineering & Sciences 2023年第10期137卷 827-841页

作者： m.B.Almatrafi abdulghani Alharbi Department of mathematics Faculty of ScienceTaibah UniversitySaudi Arabia

The extraction of traveling wave solutions for nonlinear evolution equations is a challenge in various mathematics,physics,and engineering *** article intends to analyze several traveling wave solutions for themodified regularized long-wave(mRLW)equation using several approaches,namely,the generalized algebraic method,the Jacobian elliptic functions technique,and the improved Q-expansion *** successfully obtain analytical solutions consisting of rational,trigonometric,and hyperbolic *** adaptive moving mesh technique is applied to approximate the numerical solution of the proposed *** adaptive moving mesh method evenly distributes the points on the high error *** method perfectly and strongly reduces the *** compare the constructed exact and numerical results to ensure the reliability and validity of the methods *** better understand the considered equation’s physical meaning,we present some 2D and 3D *** exact and numerical approaches are efficient,powerful,and versatile for establishing novel bright,dark,bell-kink-type,and periodic traveling wave solutions for nonlinear PDEs.

关键词： The modified regularized long wave equation soliton solutions plasma physics numerical solutions

Analytical and Numerical Investigation for the DmBBm Equation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computer modeling in Engineering & Sciences 2020年第2期122卷 743-756页

作者： abdulghani Alharbi mahmoud A.E.Abdelrahman m.B.Almatrafi Department of mathematics College of ScienceTaibah UniversityAl-Madinah Al-MunawarahSaudi Arabia. Department of mathematics Faculty of ScienceMansoura UniversityMansouraEgypt.

The nonlinear dispersive modified Benjamin-Bona-mahony(DmBBm)equation is solved numerically using adaptive moving mesh PDEs(mmPDEs)***,the exact solution of the DmBBm equation is obtained by using the extended Jacobian elliptic function expansion *** current methods give a wider applicability for handling nonlinear wave equations in engineering and mathematical *** adaptive moving mesh method is compared with exact solution by numerical examples,where the explicit solutions are *** numerical results verify the accuracy of the proposed method.

关键词： DmBBm equation Jacobian elliptic functions moving mesh PDEs(mmPDEs) moving adaptive scheme solitary waves.

Implications of Onshore Development on Global Software Engineering

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, materials & Continua 2023年第2期74卷 3029-3044页

作者： abdulrahman m.Qahtani Ahmed S.Ghiduk College of Computers and Information Technology Taif UniversityP.O.Box 11099Taif21944Saudi Arabia Department of mathematics and Computer Science Faculty of ScienceBeni-Suef UniversityBeni-Suef62521Egypt

Recently software industry has paid significant attention to customizing software products across distributed *** the requirements of multiple clients across distributed borders is a crucial challenge for the software customization *** decision-making and local development at the client site are considered methods for reducing difficulties in communicating the requirements of multiple clients across distributed *** paper introduces a new model called the onshore development model(ODm)for accomplishing the customization requests in the distributed development process of *** model presents a scenario for enhancing the onsite development of specific requirements to reduce delays andmisunderstandings between the clients and the team *** model depends on moving the development process to the client’s *** empirical studies were conducted to evaluate the proposed model to measure its productivity,time performance,and cost *** proposed model has been compared with two other models:the basic model(Bm),which allocates the decision-making process and the development process for teams at the vendor’s location,and the local decision-making model(LDec),which assigns the decision-making process for team at the client’s *** results of the empirical studies showed significant outperforming of the proposed model over the basic model and local decision-making model in productivity,time performance,and cost *** productivity of the proposed model improved by 39%and 10%more than the basic model and the local decision-making model,*** addition,the time performance of the proposed model became faster by 49%and 20.8%than the basic model and the local decision-making model,***,it reduced the total cost of the development process by 31%in terms of the salaries of all persons involved in requirements collecting,decision-making,and development.

关键词： Distributed development local development software customization

Resolution of an esophageal leak and posterior gastric wall necrosis with esophageal self-expandable metal stents

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

World Journal of Gastroenterology 2013年第40期19卷 6931-6933页

作者： majid A Almadi abdulrahman m Aljebreen Fahad Bamihriz Division of Gastroenterology King Khalid University Hospital King Saud University Riyadh 11461Saudi Arabia Division of Gastroenterology The McGill University Health Center montreal General Hospital McGill UniversityMontrealH3G 1A4Canada Division of Surgery King Khalid University HospitalKing Saud UniversityRiyadh 11461Saudi Arabia

The use of weight reduction surgeries has increased over the years with a higher proportion of these surgeries being sleeve gastrectomies,this has been associated with some complications including staple line *** report a 32-year-old male who had undergone a laparoscopic gastric band surgery and subsequently a laparoscopic sleeve gastrectomy,this was complicated by both an staple line leak at the gastroesophageal junction as well as a large(> 4 cm) posterior gastric wall defect due to gastric wall necrosis. We used two co-axially inserted self-expandable stents(SEmS) in the management of this patient,5 stents were used over repeated endoscopy sessions and 20 wk. Both defects had resolved without the need for surgical *** is the first reported case were SEmS are used for both a staple line leak as well as a gastric wall *** also review the literature on the use of SEmS in the management of leaks post weight reduction surgeries.

关键词： Stents Self-expandable metal stents Laparoscopic sleeve gastrectomy Staple line leak Esophageal leak Gastric necrosis

Sedated vs unsedated colonoscopy:A prospective study

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

World Journal of Gastroenterology 2014年第17期20卷 5113-5118页

作者： abdulrahman m Aljebreen majid A Almadi Felix W Leung Gastroenterology Division King Khalid University HospitalKing Saud UniversityRiyadh 11461Saudi Arabia Gastroenterology Division McGill University Health Center mcGill University MontrealQuebec H3G 1A4Canada Research and medical Services Sepulveda Ambulatory Care CenterVeterans Affairs Greater Los Angeles Healthcare Systemand David Geffen School of Medicine at UCLALos AngelesCaliforniaCA 91343United States

AIm:To compare sedated to unsedated colonoscopy in terms of duration,pain and the patient’s willingness to repeat the ***:Consecutive patients who underwent colonoscopies over a 2-year period were invited to *** patients who were to undergo our endoscopy unit were offered sedation with standard intravenous sedatives and analgesics,or an unsedated colonoscopy was *** details were *** patient anxiety level prior to the procedure,time to reach the cecum,total discharge time,patient and endoscopist pain assessments,satisfaction after the examination and the patient’s willingness to return for the same procedure in the future were ***:Among the 403 observed patients,more males were observed in the unsedated group(66.2%vs 55.2%,P=0.04).Additionally,the unsedated group patients were less anxious prior to the procedure(5.1vs 6.0,Pmpletion rates were comparable between the 2 groups(85.9%vs84.2%,P=0.66).The time to reach the cecum was also comparable(12.2 min vs 11.8 min);however,the total discharge times were shorter in the unsedated group(20.7 min vs 83.0 min,Pmoreover,the average patient pain score(3.4 vs 5.7,Pms of willingness to repeat the procedure if another was required in the future(83.3%vs 77.3%,P=0.17).CONCLUSION:Unsedated colonoscopy is feasible in willing *** option saves the endoscopy units up to one hour per patient and does not affect the patient willingness to return to the same physician again for additional colonoscopies if a repeated procedure is needed.

关键词： Sedation Colonoscopy Unsedated Screening Endoscopy