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 99 for “"pre-trained model"”.
-
Low-shot Learning based on Pre-trained Model
… of low-shot learning by aiming to equip learning models with the capability to generalize effectively from limited examples. It explores various low-supervision methodologies, including self-supervised, unsupervised, semi-supervised, and active learning, with the goal of reducing the reliance on …
-
Characteristics of different deep neural networks and application of pre-trained model without transfer learning
… goal of this thesis is using data simulations to present different characteristics of three deep neural networks: fully connected deep neural network, convolutional neural network, recurrent neural network, which will perform best when dealing with different feature patterns. By using these …
-
Fine-grained sentiment analysis for customer review
… provide more useful information, the proposed models' performance were relative poor compared with document-level or sentence-level sentiment analysis due to the lack of well-labeled aspect-level dataset. This thesis propose a semi-supervised approach using pre-trained BERT model to conduct the …
-
Fotokameros tipo nustatymas naudojant gilųjį mokymąsi /
… optimization functions, loss functions, the pre-trained model, the role of hyperparameters in the training of the networks, and how the appropriate hyperparameters can be found. The paper then analyses the work related to the camera type identification problem and describes what the dataset …
-
Machine Learning-Based Production Forecasting for Multiphase Reservoirs
… such as numerical simulations, reduced-order modeling, and decline curve analysis can be time-consuming and computationally challenging, especially with large and high-resolution geomodels. To address these limitations, a novel machine-learning-assisted modeling technique has been developed …
-
Automatic Classification of Voice Disorders and Phonation Types from Speech Signals
… of three phonation types (breathy, modal, and pressed). In the classification of voice disorders, the use of convolutional neural networks (CNNs) with 2- dimensional spectral feature representations achieved better performance than 1-dimensional features. Moreover, the use of DA methods in the …
-
Toward Data and Annotation Efficient Algorithms for Medical Image Analysis
… bringing obstacles for learning a omnipotent model for all interested targets. More challenges come from the industry situation. For instance, some institutions and companies publicly release their pre-trained model, but are unwilling to release their internal used datasets for privacy …
-
Deep Semi-Supervised Learning Methodologies and Applications
… a fundamental research topic that aims to build prediction models from a limited amount of labeled data and a large amount of unlabeled data. It has shown promising results in practical applications and is an effective solution to reduce the burden of annotation. In this thesis, we study deep …
-
Self-supervised pre-training for drowsiness prediction
… with machine learning emerging as a more widespread approach. However, the lack of high-quality data frequently hinders the development of neural networks, also in this domain. Over the past decade, pre-training of neural networks using self-supervised learning techniques has proven to be an …
-
Bridging the Gap: Adaptation Approaches for Under-Resourced Language Families
Multilingual large language models have demonstrated remarkable success across a variety of natural language processing (NLP) tasks. However, their performance on low and under-resourced languages remains significantly limited, primarily due to disparities in data availability. This thesis …
-
Efficient Deep Learning and Its Applications
… <p>With these concerns, many network compression approaches are developed, such as quantization, neural architecture search, network pruning, and knowledge distillation. These approaches reduce the sizes and computational costs of DNNs while maintaining their performance.</p> <p>In this …
-
Automatic Activity Identification in Unstructured Communication Artifacts
… from collaboration tools is still in the preliminary stages. Generic actions such as promises or requests are identified in most of the studies, while fine-grained actions in conversations are less focused on. The machine learning models are trained using large, labelled datasets for the …
-
On Training, Inference, and Sample Efficiencies of Language Models
Large language models have demonstrated superior performance in various natural language processing tasks such as machine translation, natural language understanding, and natural language generation. However, despite the recent developments, language models still face critical challenges. In this …
-
Manipulating Explanations: Modifying Feature Visualizatio in in Artificial Neural Networks
… has been an increasing concern with their uninterpretable nature, and a push towards stronger techniques for interpretation. Feature visualization is one of the most popular techniques to interpret the internal behavior of individual units of trained deep neural networks. Based on activation …
-
Additive logistic mechanism for privacy-preserving self-supervised learning
… attacks, especially in the stage in which a pre-trained model is fine-tuned for the actual task. We focus on self-supervised learning applied to neural networks and design a post-training privacy-protection algorithm for the neural networks. We introduce a differential privacy mechanism, …
-
Deep Learning Methods for Novel Peptide Discovery and Function Prediction
… methods for protein identification and property prediction, encompassing two primary areas: mass spectrometry-based protein sequence identification and protein property prediction. We introduce a method that enhances the identification rate of MHC-I peptides and facilitates the discovery of novel …
-
Retrieval augmented retrieval with in-context examples
… examples, widely used in decoder-only language models (LLMs), can improve embedding model performance in retrieval tasks. Unlike in LLMs, naively prepending in-context examples (query-document pairs) to the target query at inference time does not work out of the box. We introduce a simple …
-
Neural Text Generation from Structured and Unstructured Data
… papers have shown that simpler, properly tuned models are at least competitive across several natural language processing tasks. In this thesis, we show that this is also the case for text generation from structured and unstructured data. Specifically, we consider neural table-to-text generation …
-
GeoTTER: Leveraging Local Geometry of Optimal Transport for Zero-Shot Classification
We present GeoTTER, a novel framework that redefines optimal transport in the realm of zero-shot classification. Conventional methods often suffer from miscalibration and a lack of adaptability, as they rely on fixed cost matrices derived solely from pre-trained model embeddings. In contrast, …
Page 1 of 5