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 23 for “"Mixture-of-Experts"”.
-
Multi-source domain adaptation with mixture of experts
We propose a mixture-of-experts approach for unsupervised domain adaptation from multiple sources. The key idea is to explicitly capture the relationship between a target example and different source domains. This relationship, expressed by a point-to-set metric, determines how to combine …
-
A Class of Mixture of Experts Models for General Insurance Ratemaking and Reserving
Understanding the effect of policyholders' risk profile on the number and the amount of claims, as well as the dependence among different types of claims, are critical to insurance ratemaking and IBNR-type reserving. To accurately quantify such features, it is essential to develop a regression …
-
The Open Class Authorship Attribution Problem: A Comparison of Mixture-of-Experts Methods within the JGAAP Framework
… to describe, test, evaluate, and compare methods of open class attribution that utilize multiple unique closed class attributions in a voting framework. By applying statistical techniques to the proportion of closed class attributions indicating individual candidate authors, we seek to determine …
-
Nonparametric High-dimensional Models: Sparsity, Efficiency, Interpretability
… simpler base models. It examines three types of ensemble methods: additive models, tree ensembles, and mixtures of experts. Each ensemble method is characterized by a specific structure: additive models can involve base learners with single or pairwise covariates, tree ensembles use a decision …
-
TOWARDS EFFICIENT TRANSFORMER SCALING
… parameter efficiency using parameter-sharing and Mixture-of-Experts, achieving strong results in both vision and language tasks. Second, we investigate transformer configurations, finding that token-level training benefits from deeper, narrower models, while sequence-level tasks face scaling …
-
Personalized machine learning for facial expression analysis
… Project, I investigated the personalization of deep convolutional networks for facial expression analysis. While prior work focused on population-based ("one-size-fits-all") models for prediction of affective states (valence/arousal), I constructed personalized versions of these models to …
-
Three Essays Applying Dynamic Models in Economics, Finance, and Machine Learning
… methods and their application in the fields of Economics, Finance, and Machine Learning. It pulls liberally from all three. In particular, this dissertation makes repeated use of multi-state modeling frameworks popular in Economics to bring a faceted view to the underlying data and detect its …
-
Low-cost Agents with Language Perception and Dynamic Inference
… significant challenges, particularly in terms of learning and inference costs. Traditional agents often suffer from high learning expenses due to their limited ability to generalize across diverse tasks and environments. Recent advances in large language models (LLMs) have shown strong …
-
Automated Cardiac Rhythm Diagnosis for Electrophysiological Studies, an Enhanced Classifier Approach
… existing algorithms misdiagnose a proportion of arrhythmias. Studies suggested artificial intelligence (AI) classifiers are accurate using ECG and intracardiac electrogram features and reviews suggested new features might augment diagnosis. This study aimed to develop an accurate cardiac …
-
Software-Hardware Optimizations for Efficient Collective Communications in Distributed Machine Learning Platforms
… machine learning (ML) models have emerged as one of the most prominent applications in modern computing, exemplified by mixture-of-experts–based large language models. The immense resource demands of these models have driven the development of large-scale, high-performance computing platforms …
-
Load balancing and memory optimizations for expert parallel training of large language models
… machine learning tasks, but require huge amounts of computation to train and evaluate. Mixture of experts models have emerged as a way to reduce the amount of computation required for LLMs without compromising accuracy. It is necessary to distribute these large models across several devices, but …
-
OS INSPIRED COMPLETE KERNEL FUSION
… Focusing on dynamic workloads such as Mixture-of-Experts (MoE), we highlight two critical bottlenecks. First, we demonstrate how CPU-driven execution limits the exploitation of task locality and introduces artificial synchronization barriers across distributed GPU tasks, resulting in …
-
Improving Deep Learning with Probabilistic Approaches
… be used to improve deep learning?'' On the topic of uncertainty estimation, we have three sets of contributions. Firstly, we show that probabilistic inference over the depth of a neural network not only side-steps challenges involved with scaling inference to the large weight spaces of modern …
-
Behavior acquisition in artificial agents
… in robots and simulated agents has been a topic of increasing popularity throughout the last years. Despite impressive progress, autonomous behavior at a level of animals and humans are not yet replicated by machines. Especially when a complex environment demands versatile, goal-oriented …
-
Forecasting macroeconomic models with artificial neural networks : an empirical investigation into the foundation for an intelligent forecasting system
This study investigates the foundation of an intelligent system using Artificial Intelligent (AI) technologies to assist decision makers in a specific business problem, namely business forecasting. In time series and macroeconomic modelling, there are many assumptions being imposed on the behavior …
-
Approximate Message Passing for Matrix Regression
… typically require the signal to be in the form of a vector, and the design matrix to have independent and identically distributed Gaussian entries. In this thesis, we explore estimation problems where previous assumptions of AMP algorithms are no longer satisfied. To relax the vector signal …
-
Enhancing diversity in generative commonsense reasoning for explaining relationships between concepts
Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2023-09-01 without embargo terms
-
Interpretable Deep Learning for Time Series
… research efforts to improve the interpretability of deep neural networks have significantly increased in the last couple of years. Nevertheless, they are mainly applied to vision and language tasks, and their applications to time series data are relatively unexplored. This thesis aims to identify …
-
Resource-Efficient Collaborative Training and Inference of Foundation Models in Edge-AI
The convergence of Edge Artificial Intelligence (Edge-AI) and foundation models marks a transformative paradigm shift in the design of intelligent systems. Edge-AI enables computation to be performed closer to data sources and across distributed network edges, offering significant benefits in …
-
Precipitation prediction over High Mountain Asia using Gaussian processes
The rivers of High Mountain Asia provide freshwater to over 1.9 billion people. However, precipitation, the main driver of river flow, is still poorly understood due to limited direct measurements in this area. Existing tools to interpolate the few available measurements or to downscale and …
Page 1 of 2