Computer Engineering Commons^™

Gpu Parallel Acceleration Framework For Heuristic Optimization Algorithm, Dongjie Wang, Sixin Wen, Wanzhi Meng, Di Wu

Journal of System Simulation

Abstract: Heuristic optimization algorithm are a type of algorithm that uses large-scale populations for iterative calculations and are widely used to solve all kinds of complex optimization problems. However, such algorithm have the disadvantages of large calculation and long time consumption. To solve this problem, heuristic optimization algorithms are parallelized using GPU and compute unified device architecture (CUDA) to substantially improve computational efficiency. A GPU parallel framework for heuristic optimization algorithm is proposed, which designs an information interaction framework and algorithm parallel optimization strategy with a parallel logical structure, and solves the problem of the dissimilarity of the logical structure …

An Improved Cat Swarm Optimization For Heterogeneous Multiple Mobile Robots, Liang Kang, Yi Du, Lihua Yin

Journal of System Simulation

Abstract: At present, it is difficult to achieve the real homogeneity of the members of the multiple mobile robots. The existing swarm intelligence algorithm is also difficult to accommodate the heterogeneity of the team. Focusing on the heterogeneous cooperation of multiple mobile robots, the concept of mother and child robots is proposed. In order to realize the application of swarm intelligence algorithm in heterogeneous multiple mobile robots, the basic cat swarm algorithm is improved. The subdomain and neighborhood of cat swarm are defined, and eight improvements of cat swarm algorithm are proposed, including priority of search direction, extended trajectory tracking …

Digital Twin-Driven Structural Thermal Deformation Compensation System For Radio Telescopes, Zhen Lei, Yuhua Liu, Kai Ding, Haoxiang Chen, Dongwei Li

Journal of System Simulation

Abstract: The structural thermal deformation of large-scale radio telescopes under solar thermal load cannot be measured in real-time and compensated dynamically. To solve this problem, a digital twin-driven structural thermal deformation compensation method and system is studied. Based on the fusion of measured data and simulation data, a temperature field modeling method is proposed. A simulation and prediction model of structural thermal deformation is established, and a dynamic compensation model of structural thermal deformation is built. A digital twin-driven dynamic structural thermal deformation compensation system for radio telescopes is developed. A micro-experimental model is studied to verify the effectiveness of …

Research On Path Optimization Algorithm In Dynamic Routing Environment, Xin Xie, Xiaobing Hu, Hang Zhou

Journal of System Simulation

Abstract: In a real dynamic routing environment, static path optimization (SPO) and traditional dynamic path optimization (DPO) tend to encounter issues such as detours, reversals and high computational complexity due to frequent real-time optimization calculation. To address these problems, a novel restart co-evolutionary path optimization (RCEPO) method based on the ripple-spreading algorithm (RSA) is proposed. This method integrates the path optimization process with the dynamic changes of the routing network environment to enhance the effectiveness of path optimization. Moreover, the path reoptimization calculation is performed only when the dynamic changes in the routing environment exceed the predicted range, thereby reducing …

Estimation Of The Berthing Parameter Of Unmanned Surface Vessels Based On 3d Lidar, Haichao Wang, Yong Yin, Qianfeng Jing, Lin Cong

Journal of System Simulation

Abstract: Accurate estimation of berthing parameters is a prerequisite for unmanned surface vessel autonomous berthing. A method for berthing parameter estimation is proposed based on shipborne 3D LiDAR. The method consists of two main modules: ship pose estimation and berthing state estimation. In the berthing position estimation module, raw point cloud data undergoes preprocessing algorithms aims at downsampling and removing outliers. Point cloud registration algorithms are employed to determine the vessel's position during the berthing process. The berthing state estimation module extracts berth boundary information by using the MSAC algorithm, and on the basis of this information, calculates the berthing …

Faids: Artificial Intelligence Developmental Systems Framework For Predicting And Preventing Cyberattacks In Supply Chain Networks, Lordt Becklines

Research & Publications

Cyber threats and attacks disrupt and damages supply chain networks (SCNs), which are complex and interlinked. Current methods to predict and prevent cyberattacks are inadequate and ineffective. This research proposes an AI developmental systems framework (FAIDS) to protect SCNs from cyberattacks. The framework has four components: (1) an AI threat intelligence system; (2) an AI risk assessment system; (3) an AI decision support system; and (4) an AI learning and adaptation system. The framework is tested on a simulated retail SCN. The results show that the framework can predict and prevent cyberattacks and improve the network's resilience and security. The …

Synthetic Data Generation For Accurate, Fair, And Private Recommender Systems, Derek B. Lilienthal

Master's Theses

A notorious challenge for recommender systems on online platforms is to accurately and fairly recommend items that align with users’ preferences while retaining user privacy. These systems often rely on historical data that is partially labeled and includes incomplete user information, leading to the over-representation of majority groups and disproportionately favoring popular items. Substituting user data with synthetic data can address these concerns, but accurately replicating real-world datasets has been a challenging problem. Recent advancements in generative AI have demonstrated the impressive capabilities of diffusion models in generating realistic data across several domains. In this work, we introduce two variations …

Reinforcement Learning Based Planning And Control For Robotic Source Seeking Inspired By Fruit Flies, Gauravkumar Koradiya

Master's Theses

In the exploration of mobile sensing agents such as robots equipped with sensors, these agents have the potential to measure unknown scalar fields, ranging from chemical concentrations to temperature fluctuations. One of the primary interests is the robots’ ability to gravitate towards the peaks or troughs of these fields, a phenomenon termed as “source seeking”. This behavior is invaluable in practical scenarios, like detecting chemical leaks or locating survivors after catastrophes. However, the challenge becomes significantly more pronounced when operating in turbulent flow fields, where the erratic nature of odorant plumes makes predictions and simulations highly unreliable. Interestingly, fruit flies …

Controllable Physics-Informed Latent Space For Acoustic Wave Manipulation, Tristan Shah

Master's Theses

Partial Differential Equations (PDEs) are foundational descriptive tools in scientific and engineering disciplines, enabling the exploration of phenomena from material design to climate dynamics and the propagation of pandemics and wildfires. Establishing control of PDEs, especially when the underlying phenomena are only partially observable, holds great potential for fostering technological innovations. This ambition faces considerable obstacles, as a robust sample-based solution remains elusive. Addressing this challenge, this thesis introduces a framework leveraging physics-informed machine learning (ML) to enable control of PDEs. Central to this framework is an agent, equipped with sensors to perceive its environment, that generates a low-dimensional physics-informed …

An Exploration Of Dimensionality Reduction Of Dynamics On Lie Groups Via Structure-Aware Canonical Correlation Analysis, Wooyoung Chung

Master's Theses

Incorporating prior knowledge into a data-driven modeling problem can drastically improve performance, reliability, and generalization outside of the training sample. The stronger the structural properties, the more effective these improvements become. Manifolds are a powerful nonlinear generalization of Euclidean space for modeling finite dimensions. When additionally assuming that the manifold carries (Lie) group structure, this imposes a drastically stricter global constraint. The range of their applications is very wide and includes the important case of robotic tasks. We apply this idea to Canonical Correlation Analysis (CCA). In traditional CCA one constructs a hierarchical sequence of maximal correlations of up to …

Domain Specific Aspect Extraction For Product Design, Divyam Sobti

Master's Theses

As technology advances, computers become increasingly proficient at interpreting and translating human language into machine-understandable text. With the help of algorithms in natural language processing (NLP), machines can now translate textual data. These algorithms help identify and extract specific text components known as aspects. The aspects represent specific attributes or topics within textual data. For instance, an Amazon review states,“ This phone has good battery life but poor camera quality,” and attributes like ‘battery life’ and ‘camera quality’ represent aspects in the text. Aspect extraction is a pivotal process involving identifying and isolating key features or topics within text. This …

Analyzing The Usability, Performance, And Cost-Efficiency Of Deploying Ml Models On Various Cloud Computing Platforms, Hongyu Wang

Masters Theses (Archived)

With the enhanced computing capabilities and accessibility to cloud resources, major cloud computing providers such as Google Cloud Platform (GCP), Amazon Web Services (AWS), and Microsoft Azure offer Machine Learning (ML) and AI services. Their primary purpose is to provide efficiency, scalability, and adaptability in modern software development and IT operations while reducing overall costs and operational complexity. However, prospective customers of the services often question which ML-AI service will best suit their organizational and business needs. This study compares and analyzes the usability, performance, and cost-efficiency of deploying Machine Learning (ML) models across three cloud platforms: GCP, AWS, and …

Healthcare Assistance Challenges-Driven Neurosymbolic Ai, Kaushik Roy

Publications

Although Artificial Intelligence technology has proven effective in providing healthcare assistance by analyzing health data, it still falls short in supporting decision-making. This deficiency largely stems from the predominance of opaque neural networks, particularly in mental health care AI applications, which raise concerns about their unpredictable and unverifiable nature. This skepticism hinders the transition from information support to decision support. This presentation will explore neurosymbolic approaches that combine neural networks with symbolic control and verification mechanisms. These approaches aim to unlock AI’s full potential by enhancing information analysis and decision-making support for healthcare assistance.

A Personalized Chat Application For Career Profiling: A Case Study, Dhiraj Choithramani

Harrisburg University Dissertations and Theses

Artificial intelligence (AI) has rapidly transformed numerous fields over the past decade, significantly influencing industries such as software engineering and computer science. One of the most impactful developments in this area is the rise of AI-driven chat applications, which have evolved from simple, rule-based systems to sophisticated platforms capable of simulating human-like conversations. These chatbots are increasingly being utilized across various sectors, including customer service, healthcare, and education, to provide users with quick, personalized responses. This research paper presents a case study of a personalized chat application designed specifically for career profiling, leveraging advanced AI technologies to deliver contextually relevant …

Activity Map Generation And Event-Based Sensor Processing With Spiking Autoencoders And Sparse Dictionary Learning, Jack Easton

Computer Science and Engineering Theses and Dissertations

This thesis explores the potential of Spiking Neural Networks (SNNs) in processing event sensor data and generating high-fidelity activity maps. Event sensors capture asynchronous binary events with high dynamic range, but traditional processing methods often fail to leverage their advantages fully. SNNs, with their asynchronous, event-driven nature, offer a promising alternative.

A Spiking Autoencoder (SAE) was employed in this thesis to handle the stochastic and sparse event data, integrating deep dictionary learning to enhance the feature space and improve activity map quality. The encoder, modeled after the VGG network, extracts features from event streams generated by speckle patterns, which are …

A Step Towards Automated Ethical Analysis In Journalism: Measuring Llms’ Performance In Extracting Sourcing Information, Jingsen Wang

Computer Science and Engineering Master's Theses

This thesis explores the potential of Large Language Models (LLMs) in automating the extraction of sourcing information from news articles, a crucial step towards enhancing transparency and ethical analysis in journalism. We evaluate the performance of two state-of-the-art LLMs, GPT-4 and Claude 3, in identifying and categorizing various source types across four diverse news articles. The thesis employs a zero-shot learning approach with two different prompt designs, assessing the models’ ability to adapt to varying source structures and prompt instructions.

Our findings reveal that while LLMs show promise in extracting sourcing information, their performance varies significantly across different article types …

Evaluating The Role Of Data Enrichment Approaches Towards Rare Event Analysis In Manufacturing, Chathurangi Shyalika, Ruwan Wickramarachchi, Fadi El Kalach, Ramy Harik, Amit P. Sheth

Publications

Rare events are occurrences that take place with a significantly lower frequency than more common, regular events. These events can be categorized into distinct categories, from frequently rare to extremely rare, based on factors like the distribution of data and significant differences in rarity levels. In manufacturing domains, predicting such events is particularly important, as they lead to unplanned downtime, a shortening of equipment lifespans, and high energy consumption. Usually, the rarity of events is inversely correlated with the maturity of a manufacturing industry. Typically, the rarity of events affects the multivariate data generated within a manufacturing process to be …

Real-Time Gun Detection In Video Streams Using Yolo V8, Harish Kumar Reddy Kunchala

Electronic Theses, Projects, and Dissertations

In this research, we advance the domain of public safety by developing a machine learning model that utilizes the YOLO v8 architecture for real-time detection of firearms in video streams. A diverse and extensive dataset, capturing a range of firearms in varying lighting and backgrounds, was meticulously assembled and preprocessed to enhance the model's adaptability to real-world scenarios. Leveraging the YOLO v8 framework, known for its real-time object detection accuracy, the model was fine-tuned to accurately identify firearms across different shapes and orientations.

The training phase capitalized on GPU computing and transfer learning to expedite the learning process while preserving …

Exploiting Physical Side-Channel Information For Offensive And Defensive Ends, Sisheng Liang

All Dissertations

Side-channel information consists of side effects of computation that range from microarchitectural to physical phenomena. Empirical studies have demonstrated the practical exploitability of these side effects in real-world systems for malicious attacks and effective defenses. In this dissertation, we discover, analyze, and exploit certain physical side-channel information for end-to-end attacks and defense across three studies.

In the first study, we demonstrate a new DNN model extraction attack named Clairvoyance that exploits certain far-field electromagnetic signals emitted from a GPU to steal DNN models several meters away from the victim machine, even with some physical obstacles in between. Using Clairvoyance, an …

Improving Expressive Capacity Of Deep Neural Networks, Clayton Harper

Computer Science and Engineering Theses and Dissertations

Deep learning has had remarkable success in a variety of fields. However, architectures often rely on hyperparameter searches and heuristics for improved model performance. Performing hyperparameter searches is an arduous task--often time-consuming and potentially expensive to run on accelerated hardware. As a result, practitioners often rely on heuristics which may lead to sub-optimal results. In the context of deep learning, hyperparameters are set by the user prior to the training process and remain fixed. Deep learning uses gradient descent to learn complex feature representations from data, limiting human intervention. While the weights of the architecture can learn directly through data …