Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 82 for “"Dataset generation"”.
-
Towards Synthetic Dataset Generation for Semantic Segmentation Networks
… of this thesis is to converge upon a synthetic dataset generation method and testing framework that is conducive towards rapid validation of unimodal semantic segmentation network architectures. We explore multiple avenues of synthetic dataset generation. Insights gained through these …
-
Acquisition and Classification of Automobile Audio Signals for Dataset Generation
… of engine characteristics leads to cleaner dataset generation and granularity in engine health categorization.
-
Dataset generation and machine learning approaches for Android malware detection
… In addition, we propose on this basis a dataset of more than 17,000 apps for the evaluation of machine learning approaches for Android malware detection. This thesis also includes machine learning experiments using classification strategies that define the legitimate static …
-
Chord Builder: A Rapid Music Dataset Generation Tool for Automatic Chord Recognition
Η παρούσα διπλωματική εργασία εξετάζει ένα κεντρικό πρόβλημα στον τομέα της Ανάκτη- σης Μουσικής Πληροφορίας (MIR), την Αυτόματη Αναγνώριση Συγχορδιών (ACR). Αν και έχουν εξεταστεί διάφορες προσεγγίσεις, η έλλειψη επισημειωμένων μουσικών δεδομένων περιορίζει την αποτελεσματική εκπαίδευση μοντέλων, …
-
A Framework for Dataset Generation, Anomaly Detection and Cybersecurity in Industrial Cyber-Physical Systems
… and validates a unified framework for the generation of datasets in industrial CPS aimed at anomaly detection and industrial cybersecurity. To this end, a methodology is proposed that combines a hybrid and flexible experimental testbed based on real industrial technologies and digital …
-
Streaming-based CNN architecture for physical layer wireless receivers : deployment-oriented dataset generation and quantisation methods
… by the ADC, this thesis introduces the DeepRFSoC dataset generation methodology, which enables training on realistic loopback data affected by simulated channel impairments and hardware-specific distortions. This methodology enables the accelerator to operate in real-time on the SDR platform …
-
Assessing and Improving the Representation of Hydrologic Processes in Atmospheric, Ocean, and Land Modeling and Dataset Generation
… importance, its representation in modeling and dataset generation is problematic. Here, the depiction of three phenomena, ocean surface turbulent fluxes, humidity inversions, and groundwater, are assessed, and suggestions for improvements of their representations are made. First, ocean surface …
-
A Framework for Scalable Dataset Generation and Deep Learning-Based IoT Device Identification: Redefining the Future Paradigm
… model, developing realistic and balanced dataset, and enabling incremental learning to support newly added devices without full model retraining. A detailed examination of widely adopted public datasets reveals key limitations, notably complex network traffic capture setups, inconsistent …
-
MACHINE LEARNING BASED NONLINEAR CHARACTERIZATION IN HETEROGENEOUS OPTICAL NETWORKS
… models, while attractive for machine learning dataset generation, struggle when modeling the statistical characteristics of XPM induced phase noise. Investigation into the symbol pattern dependence, manifestation of artifacts in approximate models, and alterations made to each approximate model …
-
ML-based power analysis for ASIC IP development
… We establish a comprehensive flow encompassing dataset generation and power simulation. Subsequently, we develop and evaluate several ML models, categorized as architecture-based and flow-based approaches. Our findings reveal varying degrees of success among these models, with some demonstrating …
-
ON PRIVACY RISK OF RELEASING DATA AND MODELS
… three parts. In the first part, we use synthetic dataset generation techniques to address privacy risk while publishing datasets. We adapt and extend the Latent Dirichlet Allocation to generate travelling records of commuters. In the second part, we use differential privacy to provide quantifiable …
-
A Real-Time Autonomous Path Planning Framework for Space Satellites Using Improved Interfered Fluid Dynamic System (IFDS)
… train the predictive model, enabling efficient dataset generation. XGBoost, trained on this dataset, is then employed within the IFDS framework to predict optimal collision-avoidance parameters in real time. This two-step approach enables satellites to autonomously adjust trajectories, …
-
Scientific chart image recognition and interpretation
… and question answering (QA), and ground truth dataset generation for performance evaluation.
-
Addressing Cybersecurity Data & Workforce Scarcity with TROY: Testbed for Resilient Operational sYstems
… This has led researchers to create synthetic datasets using testbeds and cyber-ranges. These testbeds and cyber-ranges are used to instruct students on cybersecurity attack and defense strategies, perform realistic cybersecurity experiments, and create data to train effective machine learning …
-
A supervised machine learning-based framework to detect low-level fault injections in software systems
… software fault injections consisting of labelled dataset generation using an instruction-level software fault injection tool to simulate attack effects. The framework is implemented on two software systems and the results demonstrate its feasibility. The thesis explores system-level threat …
-
Language Models: Generator and Labeler
… such as ensuring label accuracy and preserving dataset diversity, this work can inspire further research on exploiting dataset-generation-based zero-shot learning using large pre-trained language models.
-
AI ASSEMBLY: OBJECT RECOGNITION, COMPUTER VISION, AND DIGITAL TWIN FOR MIXED REALITY ASSEMBLY
… prototypical study using Lego blocks, synthetic dataset generation, secondary case studies with 3D-printed nodes, real-time video streaming and processing, and assembly instruction generation. The AI-Assembly system can respond to real-time user changes, error detection, assembly status tracking, …
-
Structured representation learning by controlling generative models
… and structure play critical roles in image generation, 3D reconstruction and animation. In recent years, supervised algorithms have dramatically improved the accuracy of the learned correspondence. However, these approaches are expensive due to manual annotation and do not generalize well to …
-
Data curation for foundation model training
… we will examine ways to increase the quality of datasets in image-text and language domains, with a focus on dataset curation and filtering. Given the amount of readily available data on the web, these techniques can be reliably applied as methods to increase the downstream performance of models, …
Page 1 of 5