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 32 for “"Generalization Bounds"”.
-
Generalization bounds and size generalization for graph neural networks
… understanding of these models involves analyzing generalization bounds, a theoretical framework for finding the provable discrepancies between expected train and test loss. We make advancements in state-of-the-art PAC-Bayes generalization bounds for GNNs using insights from graph theory and random …
-
Generalization Bounds for Compressed Learning with Hard Support Vector Machines, and Multiclass Learning with Error Correcting Output Codes
… we provide theoretical results to show how the generalization bound changes with respect to the compression ratio used. These results allow for theoretical justifications in choosing the best compression matrix given the particular design parameters at hand. Additionally, as required for the …
-
Data-Dependent Analysis of Learning Algorithms
This thesis studies the generalization ability of machine learning algorithms in a statistical setting. It focuses on the data-dependent analysis of the generalization performance of learning algorithms in order to make full use of the potential of the actual training sample from which these …
-
Generalization in Machine Learning Through Information-Theoretic Lens
… information-theoretic framework to investigate generalization theory in machine learning, a critical area of research today. Specifically, we develop novel information-theoretic generalization bounds for machine learning algorithms. First, we apply information-theoretic analysis for models …
-
Learning in High Dimensional Spaces: Applications, Theory, and Algorithms
… algorithm is proposed based on the results on generalization bounds and our results show that this new algorithm is better than the existing SVM and boosting algorithms.
-
Noise-aware generalization bound for quantum variational circuits.
… high-dimensional data or intricate patterns. Generalization bounds offer crucial theoretical guarantees on a quantum machine learning model’s performance on previously unseen data. However, the practical implementation of quantum machine learning models in the Noisy Intermediate-Scale Quantum …
-
On Kernel-base Multi-Task Learning
… across tasks, it is possible to improve the generalization performance of each task, compared to training each individual task independently. During the past decade, most MTL research has been based on the Regularization-Loss framework due to its flexibility in specifying various types of …
-
Benefits of Additive Noise in Composing Classes of Functions with Applications to Neural Networks
… total variation and Wasserstein distances. The bounds for composition then come naturally through the use of data processing inequality. Capacity of Neural Networks: We instantiate our results for the case of sigmoid neural networks. We start by finding a bound for the single-layer noisy neural …
-
On the optimization and generalization of self-attention models : a stability and implicit bias perspective
… this success. However, the optimization and generalization dynamics of Transformer models remain poorly understood. In this thesis, we take steps towards establishing a foundational theory. Specifically, we focus on self-attention models, the core mechanism of the Transformer. We first study …
-
ON SYMMETRIC STRUCTURES IN MACHINE LEARNING: NEURAL NETWORKS, GAMES, AND GRAPHICAL MODELS
… self-attention, while symmetry enables favorable generalization bounds. Next, we study graphon mean-field games, where invariance under measurable bijections facilitates the derivation of sample complexity for learning Nash equilibria. Finally, we explore structure learning in hidden tree models, …
-
Improved Multi-Task Learning Based on Local Rademacher Analysis
… an individual subject may lead to deteriorated generalization performance. In such cases, learning multiple related tasks together might offer a better performance by allowing tasks to leverage information from each other. Multi-Task Learning (MTL) is a machine learning framework, which learns …
-
Approximating Bayes: Inference and Modeling
… theoretical tools for developing the so-called generalization bounds in machine learning theory. Through the analysis of various sources of error, this thesis makes contributions to the study of: convergence of high-dimensional MCMC; optimization of high-dimensional MCMC; Bayesian model …
-
Weaving order from uncertainty: Design, Analysis, and Applications of Transport-based Generative Models
… of generative models. We derive a priori error bounds that quantify the generalization performance of these models in terms of model and sample complexity. Building upon this theory, we introduce a flow-based transport algorithm for generative modeling that utilizes kernel methods to minimize …
-
Towards a Theory and Practice of Open-Ended Reasoning with Generative Models
… On the theoretical side, I establish novel generalization bounds on the performance of several generative model architectures in terms of model size and number of training samples. I then demonstrate these bounds can be used to understand commonly observed "scaling laws" during large model …
-
Theoretical foundations of Graph Neural Networks: from expressivity to generalization
… (1-WL) test (K. Xu et al., 2019), and (2) their generalization behavior is still poorly understood. This dissertation presents several contributions that improve our understanding of these challenges in GNNs. We begin by addressing the expressivity bottlenecks of standard GNNs. At the graph …
-
Deep Learning on Graphs: Theory, Models, Algorithms and Applications
… from data to facilitate explainability and generalization, which current deep learning systems can not parallel. In this thesis, I present our work, which revolves around how to improve deep learning for graphs from aspects of theory, models, algorithms, and applications.We first provide a …
-
Data Efficient Deep Learning for Rolling Element Bearing Fault Diagnosis
… loaded data can deliver superior cross domain generalization even with limited samples. Benchmark publicly available bearing datasets are analyzed, revealing weak cross domain coverage that restrict the transferability of ML models to industrial scenarios. To address this gap, a new bearing …
-
Information-Theoretic Measures of Generalization in Machine Learning
… challenges to deploying ML methods is ensuring generalization, or the ability of the model to perform well on unseen data as well as on training data. In this thesis, we explore generalization from an information-theoretic perspective. This perspective allows us to examine the impact of learning …
-
On the Stability of Structured Prediction
… by the stability of inference. First, I provide generalization bounds for learning from a limited number of examples with large internal structure. The effective learning rate can be significantly sharper than rates given in related work. Under certain conditions on the data distribution and …
-
On clustering, classification and dictionary learning: a theoretical revisit of robust kmeans, nearest-neighbour classification and dictionary learning with Moreau envelopes
… Euclidean squared loss. In this part, we derive generalization bounds for the dictionary learning problem when the empirical loss function is a Moreau envelope, a well studied function transformation with origins in variational analysis. A new sample complexity result, concerning the case of …
Page 1 of 2