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 39 for “"Machine Learning Training"”.
-
Towards Efficient AI for Science in Scalable and High Performance Distributed System
… natural language processing. In recent years, machine learning methods have been increasingly applied to the scientific discovery process, accelerating advances in diverse fields. Notable examples include AlphaFold, which predicts protein structures, and ClimateX, which enhances weather …
-
Machine Learning and Simulations for Physical Neural Networks
Machine learning, particularly methods based on neural networks, has gained significant attention since 2012, resulting in remarkable advancements in fields such as image recognition and natural language processing. However, the development of machine learning faces two significant challenges. The …
-
SMART FABRICATION 2.0: MACHINE LEARNING FOR DESIGNER-GUIDED ADAPTATION IN CLAY 3D PRINTING
… representation, forming customized datasets for machine learning training. The model then learns the logic of specific design intentions and fabrication modification styles through training, which informs future users and supports adaptive design interpretation. This study investigates how …
-
Generalized Arousal Prediction through Machine Learning
… of arousal is of immense importance in human-machine interactions and human-human interactions. With the advent of ubiquitous unobtrusive physiological sensing, continuous measurement of arousal in the lab and in the wild became feasible. In this work, we investigate the question if there are …
-
Exploratory machine learning strategies for predicting thermal conductivity of materials from transient plane source measurement
This study introduces the application of machine learning to the Hot Disk Transient Plane Source (TPS) method, aimed at enhancing the precision and efficiency of thermal conductivity prediction. Comprising two distinct parts, Part I, the research addresses the prediction of thermal conductivity in …
-
Combining Supervised Pretraining and Reinforcement Learning for Scalable Low-Thrust Trajectory Design
This thesis considers a hybrid machine learning training framework that combines supervised pretraining with deep reinforcement learning to approximate optimal low-thrust control for heliocentric transfer and rendezvous trajectories. The primary test case is a circular two-body transfer problem in …
-
A robotics system for large-scale plant monitoring in tomato greenhouses
… and dense data collection, and hierarchical deep learning architecture for automated analysis of plant images. The robotic platform was evaluated using teleoperated and semi-autonomous modes, and the design included features for full autonomy. The hierarchical scene understanding system …
-
Correct Programs, Executed Correctly: Verifying Specifications And Executions
… correct execution of optimization problems like machine learning training; and Zippel, a language for implementing and automatically verifying properties of non-interactive zero-knowledge protocols. Each one of those works shows that, by carefully designing languages and proof systems for …
-
Accelerating Collective Communication for Distributed Machine Learning
… has emerged as a cornerstone of distributed machine learning, enabling datacenter-scale clusters of accelerators to collaboratively train or serve large language models. However, it has also become a significant performance bottleneck, impeding the efficient utilization and scalability of …
-
Preserving Structural Alias Information Across the MLIR-to-LLVM Lowering Boundary
Machine learning training and inference workloads run at massive scale, where the efficiency of generated machine code directly affects throughput and energy consumption. The compilers that lower model specifications to hardware instructions rely on optimization passes that can only exploit …
-
Game of threads: Enabling asynchronous poisoning attacks
… sizes continue to grow at an unprecedented rate, machine learning training is being forced to adopt asynchronous training algorithms to maintain performance and scalability. In asynchronous training, many threads share and update the model in a racy fashion to avoid inter-thread synchronization. …
-
The Impact of Big Data Utilization on Quality Improvement in Inpatient Facilities
… can be analyzed and built into a tools, with machine learning, to aid in reduced readmission rates and overall positive patient outcomes. Purpose of the Study: The intention of this study was to evaluate the ways that big data can be analyzed to improve healthcare, specifically readmissions, …
-
Computational prediction of Ds transposon insertion sites in plants using DNA structural features
… and/or preparation of feature vectors for machine learning. It is developed for building a semi-automatic pipeline to explore various DNA features of any collection of genomic DNA sequences of interest and to prepare feature vectors for</p> <p>further machine learning.</p> <p>By use of …
-
Machine Learning for Efficient and Robust Datacenter Performance Management
… and address these challenges utilizing practical machine learning frameworks. </p><p>First, I will present my work on tuning complex configuration parameters for datacenter applications. Configuration parameters have a significant impact on application execution, and finding the optimal …
-
Sparse Bayesian reinforcement learning
… and retention methods for efficient and robust learning. We propose a framework for learning and memorizing, and we examine how we can use the memory for efficient machine learning. Temporal difference (TD) learning is a core part of reinforcement learning, and it requires function …
-
Machine learning based model predictive control with koopman linearization
… model accuracy is only guaranteed within the training frequency and/or amplitude ranges and the PEA dynamics is sensitive to operation environment (e.g., temperature and humidity), efficacy of this technique is limited to the LSTM training bandwidth and can be downgraded when the PEA dynamics …
-
The Flexible Face: Unifying the Protocols of Facial Recognition Technologies
… overtly present in the facial databases used in machine learning training and operationalizing of the technology, act under predictive logics to categorize and hierarchize the faces under observation into stable data defined by difference; political protocols, in combinations of state and …
-
Optimizing Distributed Computing Systems via Machine Learning
… algorithms. In this dissertation, we apply machine learning techniques to optimize distributed systems. Particularly, we contribute to the following two areas that drive the advance of distributed systems in resource provisioning, task scheduling, and architecture design. Optimizing …
-
Flexible and Efficient Application-Networking Co-Design in Cloud Datacenters
… and data-intensive appli- cations, from machine learning training to large-scale analytics and microservices. However, the networking abstractions available to these applications remain largely unchanged, con- tinuing to expose a low-level, byte-stream interface that lacks awareness of …
-
Trustworthy Machine Learning with Data in the Wild
Recent advances in machine learning (ML) have been largely fueled by models trained on extensive internet-collected datasets. While this approach has yielded remarkable capabilities, it introduces critical vulnerabilities: training data can be untrustworthy, containing harmful content or becoming …
Page 1 of 2