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 17 of 17 for “"Backdoor Attack"”.
-
Backdoor in AI: Algorithms, Attacks, and Defenses
… Among the various threats to these systems, backdoor attacks pose a particularly insidious challenge. These attacks compromise the model by embedding a hidden backdoor function, which can be triggered by specific inputs to manipulate the model's behavior. My research goal initially involves …
-
On the Security and Adaptation of Neural Networks: A Study of Adversarial Robustness, Backdoor Attacks, and Transfer Learning
… models are vulnerable to various adversarial attacks, including evasion, data poisoning, and backdoor attacks, which can compromise their performance and reliability. On the other hand, when these models are deployed as prebuilt solutions, pretrained on tasks different from their intended …
-
Towards the Safety and Robustness of Deep Models
… resilience against external threats, such as backdoor attacks. Consequently, this dissertation explores the implications of robust representation learning in three principal areas: i) Backdoor Attack, ii) Backdoor Defense, and iii) Noisy Labels. First, we study the backdoor attack creation and …
-
Robust and hardware efficient hardware accelerator design for convolutional neural networks
… associated with hardware Trojans (HTs) and backdoor attacks. A comprehensive literature review highlights the need to mitigate these threats, as backdoors attacks can subtly alter classifications, and HTs can cause targeted errors. Meanwhile, the increasing computational demands of CNNs and …
-
Towards More Realistic and Practical Graph Backdoor Attacks
… shown that they are highly vulnerable to graph backdoor attacks (GBAs), where adversaries implant malicious triggers to induce targeted misclassification during inference. Despite their effectiveness, existing GBAs are often developed under unrealistic assumptions, such as focusing exclusively …
-
ON ADVERSARIAL MACHINE LEARNING AND ROBUST OPTIMIZATION
… first introduce an understanding of adversarial attacks from the view of robust optimization, including the relation between commonly used uncertainty set and several types of adversarial samples, adversarial training and min-max framework, and Wasserstein adversarial set and distributionally …
-
Robust learning of deep neural networks under data corruption
… learning algorithm for corrupted supervision and backdoor attack. Specifically, in Chapter 2, I proposed the Robust Collaborative Autoencoder (RCA) approach to enhance the robustness of vanilla autoencoder methods against natural corruption. In Chapter 3, I developed Robust RealNVP, a robust …
-
Machine Learning in Industry: Applications and Security Challenges
… security challenges such as data poisoning attacks, adversarial attacks, model stealing attacks, and membership inference attacks. To address these concerns, we investigate the reliability and robustness of the crack detection system. Specifically, we design a backdoor attack, a type of data …
-
Adversarial Attack and Defense in Time Series Classification
… of security threats, including adversarial and backdoor attacks, and face limitations in certified defense and architectural robustness. This thesis addresses four interrelated challenges in robust and secure TSC, with each contributing to a deeper understanding of vulnerabilities and solutions …
-
A Multi-Domain Trojan Detector for Deep Neural Networks
Backdoor attacks have been demonstrated to compromise the functioning of machine learning models that utilize deep neural networks (DNNs). An adversary carrying out a backdoor attack embeds a predefined perturbation called a Trojan trigger into a small subset of input samples. The DNN can then be …
-
Backdoors in contemporary artificial intelligence: the good and the bad
… AI techniques are vulnerable to adversarial attacks, raising significant safety concerns about their use. Backdoor attacks have gained particular attention among various adversarial attacks against AI due to their stealthy nature. A backdoor attack refers to a generic class of attacks (from …
-
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
-
SEDPD: Sampling-Enhanced Differentially Private Defense against Backdoor Poisoning Attacks of Image Classification
… decision processes through explanation-guided attacks. In this paper, we propose a robust, model-agnostic defense framework to mitigate these vulnerabilities by utilizing explanation-values from XAI-frameworks while preserving the utility of explanation-values. Our framework employs a …
-
Certified and Forensic Defenses against Poisoning and Backdoor Attacks
Data poisoning and backdoor attacks manipulate model predictions by inserting malicious instances into the training set. Most existing defenses against poisoning and backdoor attacks are empirical and easily evaded by an adaptive attacker. In addition, existing empirical defenses provide, at best, …
-
Training Phase Specific Backdoor Defences For Federated Learning
… devices. Such manipulations can lead to backdoor attacks that embed undesirable behaviors into the joint model. This serious security issue needs to be addressed before FL systems can be deployed into critical infrastructure. This thesis proposes strategies to protect FL systems from …
-
Trustworthy machine learning throughout model’s life cycle
Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2023-09-01 without embargo terms
-
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 …