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 16 of 16 for “"distributional robustness"”.
-
Distributional robustness: towards challenging and real-world settings of distribution shift
<p>The ever-evolving world we live in causes continuous data distribution shifts for any machine learning application. As machine learning models are notoriously not robust against such distribution shifts, it is of our interest to develop techniques to deal with this problem. <strong>Hence, this …
-
COMPLETELY POSITIVE AND CO-POSITIVE PROGRAMMING: APPLICATIONS IN DISRUPTION RISK MANAGEMENT, POST DISASTER HUMANITARIAN LOGISTICS AND HOMELAND SECURITY GAMES
… Specifically, by adopting the concept of distributional robustness and assuming the first-two moments of the underlying distribution are known, they showed the moment-based bound of a mixed 0-1 linear problem with random objective coefficients can be obtained by solving a completely …
-
Model-Based and Data-Driven Covariance Control: Theory and Applications
… methods often struggle with ensuring safety and robustness in stochastic environments. This dissertation advances the theory of covariance steering (CS), which shifts the focus from controlling specific system states to steering entire state distributions under constraints. Historically, CS has …
-
Randomized dimension reduction with statistical guarantees
… augmentation that improve generalization and distributional robustness provably. Specifically, Chapter 4 presents a sample complexity analysis for data augmentation consistency regularization where we view sample efficiency from the lens of dimension reduction in the function space. Then in …
-
Learning-based intelligent control and safety assurance of unmanned autonomous vehicles
… demands for multi-agent systems; (iii) robustness—resilience to uncertainty including non-Gaussian disturbances.<br/><br/>A data-driven approach identifies robot dynamics via Sparse Identification of Nonlinear Dynamics (SINDY), superior to NARX in noise robustness. Control Barrier …
-
A Systems Approach to the Modeling and Control of Molecular, Microparticle, and Biological Distributions
… optimal control formulation, and worst-case and distributional robustness analysis. Free radical bulk polymerization is the model system for molecular distributions. In situ ATR-FTIR spectroscopy was used to determine monomer concentration; off-line gel permeation chromatography was used to …
-
A Theory of Generative Models and Robustness via Regularization
… privacy. We then focus on machine learning robustness, where one is interested in understanding the susceptibility of a model in the face of adversarial threats. We show that regularization is intimately connected to distributional robustness, which subsumes existing results and extends them …
-
Data-driven Distributionally Robust Stochastic Optimization via Wasserstein Distance with Applications to Portfolio Risk Management and Inventory Control
… dissertation is stochastic optimization under distributional ambiguity. One canthink of this as a two player game between a decision maker, who tries to minimize some loss or maximize some reward, and an adversarial agent that chooses the worst case, or least favorable, distribution (to the …
-
Machine Learning Toward Materials Discovery: From Crystal Mapping and Out-of-Distribution Property Prediction to Radiation Detection and Emerging Foundation Models.
… machine learning frameworks to improve robustness, generalization, and interpretability in materials informatics and radiation detection. The work spans crystal structure mapping, adversarial learning for out-of- distribution prediction, foundation-model-based property prediction, and …
-
Towards reliability and interactive debugging for large language models
… predictions and are robust to data biases or distributional shifts. They also want to be equipped to interact with these models to better understand and debug them. We present a variety of training and inference techniques toward building these aspects of reliability into models. We …
-
Distributionally robust binary classifier under Wasserstein distance
The robustification of statistical models has been a popular topic for decades. Statistical robustification and robust optimization are the two main approaches in the literature, where the former stabilizes the model output by removing the outlier points while the latter concerns more the outlier …
-
Theories and Algorithms for Efficient and Robust Sequential Decision Making
… (3) learned policies often exhibit brittle robustness, with significant performance degradation under slight environment shift. These obstacles have collectively prevented the widespread deployment of reinforcement learning, particularly in resource-scarce and high-stakes domains. This …
-
Certifying robustness in inference and learning problems
… answered. In this dissertation, a few topics in robustness of inference and learning methods are studied. One of the key issues with data-driven methods in practice is unexpected changes that can occur at inference time potentially affecting the performance of these methods, for instance, …
-
Robust Fragmentation: A Data-Driven Approach to Decision-Making Under Distributional Ambiguity
… an unknown distribution leading to the term distributional ambiguity. The decision maker may only have a sample of past events to guide them in balancing the costs and benefits of every possibility. Such limited historical data can help to provide guidance, but how best to use it? One would …
-
Estimating selected disaggregated socio-economic indicators using small area estimation techniques
… the distribution of the random effects and distributional robustness is automatically achieved. The MSE is estimated using an analytical approach based on Taylor series linearization. Chapter 3 presents the estimation of overweight prevalence at the county level in Kenya. In this …