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 39 for “"Adversarial Example"”.

  1. Adversarial Example Resistant Hyperparameters and Deep Learning Networks

    … systems resulting in the phenomenon of adversarial examples. Hyperparameters, the settings used to build and train machine learning models, have been shown to build machine learning models that are more resistant to adversarial examples. In this paper, we expand the research of …

    washington Repository record for Adversarial Example Resistant Hyperparameters and Deep Learning Networks (opens in a new tab)

  2. DETECTING AND DEFENDING AGAINST DIFFERENT FAMILIES OF ADVERSARIAL EXAMPLE ATTACKS

    Adversarial example attacks alter an image so the image appears largely unaltered to human eyes, but image-recognition models will misclassify it. This is a common type of attack, against which there is currently no good general defense. Most state-of-the-art methods of detecting adversarial …

    nps Repository record for DETECTING AND DEFENDING AGAINST DIFFERENT FAMILIES OF ADVERSARIAL EXAMPLE ATTACKS (opens in a new tab)

  3. FORMING ADVERSARIAL EXAMPLE ATTACKS AGAINST DEEP NEURAL NETWORKS WITH REINFORCEMENT LEARNING

    … has demonstrated that DNNs remain vulnerable to adversarial examples. While many defense schemes have been proposed to counter the equally many attack vectors, none have been successful at securing a DNN from this vulnerability. Novel attacks expose blind spots unique to a network’s defense, …

    nps Repository record for FORMING ADVERSARIAL EXAMPLE ATTACKS AGAINST DEEP NEURAL NETWORKS WITH REINFORCEMENT LEARNING (opens in a new tab)

  4. Efficient Gan-Based Adversarial Example Generation Against Ml-Based Network Intrusion Detection Systems

    … advancements, ML-based NIDS remain vulnerable to adversarial examples—deliberately crafted inputs designed to mislead models and trigger incorrect classifications. Originally identified in the field of computer vision, adversarial examples now pose a critical threat to the reliability and …

    calpoly Repository record for Efficient Gan-Based Adversarial Example Generation Against Ml-Based Network Intrusion Detection Systems (opens in a new tab)

  5. Analyzing Adversarial Examples: A Framework to Study Adversary Knowledge

    Adversarial examples are malicious inputs to trained machine learning models supplied to trigger a misclassification. This type of attack has been studied for close to a decade, and we find that there is a lack of study and formalization of adversary knowledge when mounting attacks. This has …

    waterloo Repository record for Analyzing Adversarial Examples: A Framework to Study Adversary Knowledge (opens in a new tab)

  6. DEFENDING AGAINST DEEP LEARNING-BASED VIDEO FINGERPRINTING ATTACKS WITH ADVERSARIAL EXAMPLES

    … New defenses that are robust against DNNs use adversarial examples to fool the classifier, but those defenses either assume the user has access to the full traffic trace beforehand or require expensive maintenance from Tor servers. In this thesis, we propose Prism, a defense against …

    nps Repository record for DEFENDING AGAINST DEEP LEARNING-BASED VIDEO FINGERPRINTING ATTACKS WITH ADVERSARIAL EXAMPLES (opens in a new tab)

  7. On Transferability of Adversarial Examples on Machine-Learning-Based Malware Classifiers

    … vulnerable and sensible to transferable adversarial example (AE) attacks. The transfer AE attack does not require extra information from the victim model such as gradient information. Researchers explore mainly 2 lines of transfer-based adversarial example attacks: ensemble models and …

    vt Repository record for On Transferability of Adversarial Examples on Machine-Learning-Based Malware Classifiers (opens in a new tab)

  8. Understanding Adversarial Training: Improve Image Recognition Accuracy of Convolution Neural Network

    … neural networks, consistently misclassify adversarial examples—inputs formed by applying small but intentionally worst-case perturbations to examples from the dataset, such that the perturbed input results in the model outputting an incorrect answer with high confidence. The main purpose of …

    cuny Repository record for Understanding Adversarial Training: Improve Image Recognition Accuracy of Convolution Neural Network (opens in a new tab)

  9. A* ATTACK: A NOVEL PATH-FINDING APPROACH TO ADVERSARIAL EXAMPLES

    … vulnerability of deep neural networks (DNNs) to adversarial examples with a focus on the black-box machine learning as a service (MLaaS) environment. We introduce A* Attack, a unique adversarial example attack that leverages the A* Search algorithm to find adversarial perturbations. This …

    nps Repository record for A* ATTACK: A NOVEL PATH-FINDING APPROACH TO ADVERSARIAL EXAMPLES (opens in a new tab)

  10. Reverse Engineering of Adversarial Samples by Leveraging Patterns left by the Attacker

    Intrinsic susceptibility of deep learning to adversarial examples has led to a plethora of attack techniques with a common broad objective of fooling deep models. However, we find slight compositional differences between the algorithms achieving this objective. These differences leave traces that …

    ucf Repository record for Reverse Engineering of Adversarial Samples by Leveraging Patterns left by the Attacker (opens in a new tab)

  11. ATTACKING MULTI-LABEL AUDIO CLASSIFICATION WITH GREEDY TREE SEARCH

    Adversarial example attacks provide a means to fool deep neural networks (DNNs) into making inaccurate predictions with minimal input modifications. Some attacks specialize in attacking audio, but very few attacks have been developed for use against multi-label classification. We propose modifying …

    nps Repository record for ATTACKING MULTI-LABEL AUDIO CLASSIFICATION WITH GREEDY TREE SEARCH (opens in a new tab)

  12. An Ensemble Approach for Explanation-based Adversarial Detection

    … Deep Neural Networks (DNNs) to be vulnerable to adversarial examples that induce desired misclassifications in the models. Such risks impede the application of machine learning in security-sensitive domains. Several defense methods have been proposed against adversarial attacks to detect …

    tamu Repository record for An Ensemble Approach for Explanation-based Adversarial Detection (opens in a new tab)

  13. Advancing Adversarial Audio: Human-in-the-Loop Black-box Attacks

    Adversarial audio attacks pose significant security challenges to real-world audio applications. Attackers may manipulate speech to impersonate a speaker, gaining access to smart devices like Amazon Echo. In audio applications, there are two key areas: music and speech. In music, most attackers …

    usf Repository record for Advancing Adversarial Audio: Human-in-the-Loop Black-box Attacks (opens in a new tab)

  14. Robustness evaluation of deep neural networks with provable guarantees

    … if exceeding the radius, there must exist an adversarial example. We extend the maximum safe radius to two variants: the expected maximum safe radius to evaluate global robustness of a dataset, and the maximum safe radius w.r.t. optical flow when the input is a video. We also study the feature …

    oxford Repository record for Robustness evaluation of deep neural networks with provable guarantees (opens in a new tab)

  15. Building trustworthy machine learning systems in adversarial environments

    … interacting with a machine-learning model. These adversarial scenarios, known as poisoning attack, adversarial example attack, and inference attack, have demonstrated that security, privacy, and robustness have become more important than ever for AI to gain wider adoption and societal trust. To …

    vt Repository record for Building trustworthy machine learning systems in adversarial environments (opens in a new tab)

  16. Multi-model-based defense against adversarial examples for neural networks

    … models have been found to be vulnerable to adversarial examples, i.e., carefully crafted examples aiming to mislead machine learning models. Adversarial examples can pose potential risks on safety/security-critical applications. Existing defense approaches are still vulnerable to emerging …

    uiuc Repository record for Multi-model-based defense against adversarial examples for neural networks (opens in a new tab)

  17. A Watermarking-Based Framework for Protecting Deep Image Classifiers Against Adversarial Attacks

    … tasks, they are known to be vulnerable to adversarial examples---inputs with imperceptible, but subtly crafted perturbation which fool the models to produce incorrect outputs. To distinguish adversarial examples from benign images, in this thesis, we propose a novel watermarking-based …

    waterloo Repository record for A Watermarking-Based Framework for Protecting Deep Image Classifiers Against Adversarial Attacks (opens in a new tab)

  18. An investigation of machine learning systems security with a forensics perspective

    … frameworks have been proposed to attribute adversarial examples to the models that generated them. However, the robustness of these attribution mechanisms against evasion attempts remains underexplored. This thesis investigates whether a simple stochastic perturbation strategy can undermine …

    qu-belfast Repository record for An investigation of machine learning systems security with a forensics perspective (opens in a new tab) Download PDF for An investigation of machine learning systems security with a forensics perspective (opens in a new tab)

  19. Adversarial Transferability and Generalization in Robust Deep Learning

    … (DL) models exhibit significant fragility to adversarial examples, i.e., subtle modifications applied to inputs during testing yet effective in misleading DL models. These meticulously crafted perturbations possess the remarkable property of transferability: an adversarial example that …

    must-thes Repository record for Adversarial Transferability and Generalization in Robust Deep Learning (opens in a new tab)

  20. An evaluation of the robustness of the natural-adversarial mutual information-based defense and malware classification against adversarial attacks for deep learning

    … way. These modified samples are known as adversarial examples and have been crafted with the goal of causing the target DNN to modify its behavior. It has been shown that adversarial examples can be crafted even when the attacker does not have access to the training parameters and model …

    utc Repository record for An evaluation of the robustness of the natural-adversarial mutual information-based defense and malware classification against adversarial attacks for deep learning (opens in a new tab)

Page 1 of 2