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 12 of 12 for “"Backdoor Attacks"”.
-
TextGuard: Provable defense against backdoor attacks on text classification
Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2025-05-01
-
Towards machine learning models robust to adversarial examples and backdoor attacks
… of such systems: adversarial examples and backdoor data poisoning attacks. Specifically, in the first part of the thesis, we build a methodology for defending against adversarial examples that is the first one to provide non-trivial adversarial robustness against an adaptive adversary. In …
-
Exploring the landscape of backdoor attacks on deep neural network models
… recently been demonstrated to be vulnerable to backdoor attacks. Specifically, by introducing a small set of training inputs, an adversary is able to plant a backdoor in the trained model that enables them to fully control the model's behavior during inference. In this thesis, the landscape of …
-
Beyond the Security of Deep Learning: An Exploration of Stealthy Backdoor Attacks in Computer Vision
Contains fulltext : 319701.pdf (Publisher’s version ) (Open Access)
-
Understanding and Mitigating Data-Centric Vulnerabilities in Modern AI Systems
… vulnerabilities, including data poisoning, backdoor attacks, and subtle human-exploitation vectors. This thesis addresses these challenges through a comprehensive data-centric perspective on AI security. First, we examine backdoor attacks in the frequency domain, revealing that many triggers …
-
Trojan Detection in Large Language Models of Code
… models of code (Code-LLMs) against trojan attacks. Code-LLMs are widely used in software development for tasks such as vulnerability detection, clone detection, code completion, and code summarization. Popular platforms that use AI-assisted software development include GitHub’s Copilot and …
-
Trustworthy Federated Learning Systems: From Secure Distributed Training to Reliable Fine-tuning
… updates. This design enables the detection of backdoor attacks that are difficult to identify using conventional parameter-distance methods and improves the robustness of federated model adaptation. Third, it develops Behavioral Hard Probability Optimization (BHPO), a preference-based …
-
Evolving Threats and Defenses in Machine Learning: Focus on Model Inversion and Beyond
… various emerging threats, model inversion (MI) attacks stand out due to their potential to compromise the confidentiality of training data. This dissertation investigates evolving threats in ML, centering on model inversion and its implications across image classification and natural language …
-
How Data Drives ML Models Performance
… and consider two different threat models: backdoor attacks and malicious data editing. In this first threat model, an adversary injects a few doctered samples into the training set to control model predictions at inference time. We study the effect of these malicious samples on model …
-
Using Public and Private Blockchains for Secure Data Sharing and Analytics
… hybrid blockchain architecture that discourages backdoor attacks by detecting and punishing the attackers. To further secure shared data, we show how we can use the Trusted Execution Environ- ments(TEEs) to enable efficient, privacy enhancing and secure applications by implementing oblivious …
-
Defending Against Trojan Attacks on Neural Network-based Language Models
Backdoor (Trojan) attacks are a major threat to the security of deep neural network (DNN) models. They are created by an attacker who adds a certain pattern to a portion of given training dataset, causing the DNN model to misclassify any inputs that contain the pattern. These infected classifiers …
-
Towards Robust Deep Neural Networks
… tasks. Unfortunately, they are vulnerable to attacks, such as Trojans during training and Adversarial Examples at test time. Adversarial Examples are inputs with carefully crafted perturbations added to benign samples. In the Computer Vision domain, while the perturbations being imperceptible …