<p dir="ltr">Neural representations are high-dimensional embeddings generated during the feed-forward process of neural networks. These embeddings compress raw input information and extract abstract features beneficial for downstream tasks. However, effectively utilizing these representations poses challenges due to their inherent complexity. This complexity arises from the non-linear relationship between inputs and neural representations, as well as the diversity of the learning process.</p><p dir="ltr">In this thesis, we propose effective methods to utilize neural representations for adversarial attack and defense. Our approach generally involves decomposing complex neural representations into smaller, more analyzable parts. We also seek general patterns emerging during learning to better understand the semantic meaning associated with neural representations.</p><p dir="ltr">We demonstrate that formalizing neural representations can reveal models' weaknesses and aid in defending against poison attacks. Specifically, we define a new type of adversarial attack using neural style, a special component of neural representation. This new attack uncovers novel aspects of the models' vulnerabilities. </p><p dir="ltr">Furthermore, we develop an interpretation of neural representations by approximating their marginal distribution, treating intermediate neurons as feature indicators. By properly harnessing these rich feature indicators, we address scalability and imperceptibility issues related to pixel-wise bounds.</p><p dir="ltr">Finally, we discover that neural representations contain crucial information about how neural networks make decisions. Leveraging the general patterns in neural representations, we design algorithms to remove unwanted and harmful functionalities from neural networks, thereby mitigating poison attacks.</p>

Adversarial examples are among the most widespread attacks in adversarial machine learning. In this work, we define new targeted and non-targeted attacks that are computationally less expensive than standard adversarial attacks. Besides practical purposes in some scenarios, these attacks can improve our understanding of the robustness of machine learning models. Moreover, we introduce a new training scheme to improve the performance of pre-trained neural networks and defend against our attacks. We examine the differences between our method, standard training, and standard adversarial training on pre-trained models. We find that our method protects the networks better against our attacks. Furthermore, unlike usual adversarial training, which reduces standard accuracy when applied to previously trained networks, our method maintains and sometimes even improves standard accuracy.

<p>Computer vision applications such as image classification and object detection often suffer from adversarial examples. For example, adding a small amount of noise to input images can trick the model into misclassification. Over the years, many defense mechanisms have been proposed, and different researchers have made seemingly contradictory claims on their effectiveness. This dissertation first presents an analysis of possible adversarial models and proposes an evaluation framework for comparing different more powerful and realistic adversary strategies. Then, this dissertation proposes two randomness-based defense mechanisms Random Spiking (RS) and MoNet to improve the robustness of image classifiers. Random Spiking generalizes dropout and introduces random noises in the training process in a controlled manner. MoNet uses the combination of secret randomness and Floyd-Steinberg dithering. Specifically, input images are first processed using Floyd-Steinberg dithering to reduce their color depth, and then the pixels are encrypted using the AES block cipher under a secret, random key. Evaluations under our proposed framework suggest RS and MoNet deliver better protection against adversarial examples than many existing schemes. Notably, MoNet significantly improves the resilience against transferability of adversarial examples, at the cost of a small drop in prediction accuracy. Furthermore, we extend the usage of MoNet to the object detection network and use it to align with model ensemble strategies (Affirmative and WBF (weighted fusion boxes)) and Test Time Augmentation (TTA). We call such a strategy 3MIX. Evaluations found that 3Mix can significantly improve the mean average precision (mAP) on both benign inputs and adversarial examples. In addition, 3Mix is a lightweight approach to migrate the adversarial examples without training new models.</p>

No description available.

La seguridad informática siempre ha ido asociada a los avances tecnológicos, desde los años 60 protegiendo los sistemas de forma física impidiendo su acceso hasta nuestros días en los que se utiliza la inteligencia artificial para detectar comportamientos anómalos en redes, detectar malware o dar soporte a sistemas de acceso restringido. Según Cybercrime Magazine, el impacto económico que tendrá el cibercrimen en el mundo en cinco años (2020-2025) será de aproximadamente 10 billones de dólares anuales aumentando un 15% anual. Así mismo, se espera un gasto en defensa mayor con un incremento anual de un 13 %. La IA ha tomado un papel fundamental en la detección, protección y predicción de incidentes a medida que los modelos han ido mejorando y la potencia de sistemas de cómputo han permitido un volumen mayor de datos para el entrenamiento. Actualmente la inteligencia artificial está logrando resultados sorprendentes, podemos ver como ChatGPT es capaz de generar textos, responder a preguntas complejas o generar código de programación. Dall E 2 genera imágenes de alta resolución respondiendo a complejos prompts, desde dibujos animados a todo tipo de representaciones realistas. Existen otros tipos de modelos, quizás menos mediáticos, que también han alcanzado resultados notables. Los modelos de clasificación de imágenes, detección de malware, detección de objetos o identificación biométrica son algunos ejemplos. Si bien es cierto que la IA no está exenta de polémicas relacionadas con la protección de datos o sobre su uso ético. Numerosos países y organizaciones han mostrado sus preocupaciones de cómo estas tecnologías pueden afectar a las sociedades debido a los sesgos que introducen y sobre cómo pueden impactar en los derechos humanos. En este contexto de IA y seguridad se desarrolla esta tesis. Los modelos de inteligencia artificial son vulnerables a distintos tipos ataques como Data Poisoning, Model Stealing, Model Inversion o Adversarial Attacks. En este trabajo nos enfocamos en analizar las vulnerabilidades presentes en los modelos de clasificación de imágenes y concretamente en los ataques adversarios. Dichos ataques fueron descubiertos en 2013 y desde entonces se han desarrollado múltiples técnicas y tipos de algoritmos, así como defensas o métodos para crear modelos más robustos y resistentes. En esta tesis analizamos la capacidad de los ataques adversarios para afectar a un modelo de inteligencia artificial específico y luego transferir ese conocimiento para atacar con éxito a otros modelos similares.

abstract: Machine learning (ML) and deep neural networks (DNNs) have achieved great success in a variety of application domains, however, despite significant effort to make these networks robust, they remain vulnerable to adversarial attacks in which input that is perceptually indistinguishable from natural data can be erroneously classified with high prediction confidence. Works on defending against adversarial examples can be broadly classified as correcting or detecting, which aim, respectively at negating the effects of the attack and correctly classifying the input, or detecting and rejecting the input as adversarial. In this work, a new approach for detecting adversarial examples is proposed. The approach takes advantage of the robustness of natural images to noise. As noise is added to a natural image, the prediction probability of its true class drops, but the drop is not sudden or precipitous. The same seems to not hold for adversarial examples. In other word, the stress response profile for natural images seems different from that of adversarial examples, which could be detected by their stress response profile. An evaluation of this approach for detecting adversarial examples is performed on the MNIST, CIFAR-10 and ImageNet datasets. Experimental data shows that this approach is effective at detecting some adversarial examples on small scaled simple content images and with little sacrifice on benign accuracy. / Dissertation/Thesis / Masters Thesis Computer Science 2019

There has been tremendous progress in machine learning and specifically deep learning in the last few decades. However, due to some inherent nature of deep neural networks, many questions regarding explainability and robustness still remain open. More specifically, as deep learning models are shown to be brittle against malicious changes, when do the models fail and how can we construct a more robust model against these types of attacks are of high interest. This work tries to answer some of the questions regarding explainability and robustness of deep learning by tackling the problem at four different topics. First, real world datasets often contain noise which can badly impact classification model performance. Furthermore, adversarial noise can be crafted to alter classification results. Geometric multi-resolution analysis (GMRA) is capable of capturing and recovering manifolds while preserving geomtric features. We showed that GMRA can be applied to retrieve low dimension representation, which is more robust to noise and simplify classification models. Secondly, I showed that adversarial defense in the image domain can be partially achieved without knowing the specific attacking method by employing preprocessing model trained with the task of denoising. Next, I tackle the problem of adversarial generation in the text domain within the context of real world applications. I devised a new method of crafting adversarial text by using filtered unlabeled data, which is usually more abundant compared to labeled data. Experimental results showed that the new method created more natural and relevant adversarial texts compared with current state of the art methods. Lastly, I presented my work in referring expression generation aiming at creating a more explainable natural language model. The proposed method decomposes the referring expression generation task into two subtasks and experimental results showed that generated expressions are more comprehensive to human readers. I hope that all the approaches proposed here can help further our understanding of the explainability and robustness deep learning models.

Existing adversarial algorithms for Deep Reinforcement Learning (DRL) have largely focused on identifying an optimal time to attack a DRL agent. However, little work has been explored in injecting efficient adversarial perturbations in DRL environments. We propose a suite of novel DRL adversarial attacks, called ACADIA, representing AttaCks Against Deep reInforcement leArning. ACADIA provides a set of efficient and robust perturbation-based adversarial attacks to disturb the DRL agent's decision-making based on novel combinations of techniques utilizing momentum, ADAM optimizer (i.e., Root Mean Square Propagation or RMSProp), and initial randomization. These kinds of DRL attacks with novel integration of such techniques have not been studied in the existing Deep Neural Networks (DNNs) and DRL research. We consider two well-known DRL algorithms, Deep-Q Learning Network (DQN) and Proximal Policy Optimization (PPO), under Atari games and MuJoCo where both targeted and non-targeted attacks are considered with or without the state-of-the-art defenses in DRL (i.e., RADIAL and ATLA). Our results demonstrate that the proposed ACADIA outperforms existing gradient-based counterparts under a wide range of experimental settings. ACADIA is nine times faster than the state-of-the-art Carlini and Wagner (CW) method with better performance under defenses of DRL. / Master of Science / Artificial Intelligence (AI) techniques such as Deep Neural Networks (DNN) and Deep Reinforcement Learning (DRL) are prone to adversarial attacks. For example, a perturbed stop sign can force a self-driving car's AI algorithm to increase the speed rather than stop the vehicle. There has been little work developing attacks and defenses against DRL. In DRL, a DNN-based policy decides to take an action based on the observation of the environment and gets the reward in feedback for its improvements. We perturb that observation to attack the DRL agent. There are two main aspects to developing an attack on DRL. One aspect is to identify an optimal time to attack (when-to-attack?). The second aspect is to identify an efficient method to attack (how-to-attack?). To answer the second aspect, we propose a suite of novel DRL adversarial attacks, called ACADIA, representing AttaCks Against Deep reInforcement leArning. We consider two well-known DRL algorithms, Deep-Q Learning Network (DQN) and Proximal Policy Optimization (PPO), under DRL environments of Atari games and MuJoCo where both targeted and non-targeted attacks are considered with or without state-of-the-art defenses. Our results demonstrate that the proposed ACADIA outperforms state-of-the-art perturbation methods under a wide range of experimental settings. ACADIA is nine times faster than the state-of-the-art Carlini and Wagner (CW) method with better performance under the defenses of DRL.

<p dir="ltr">This research explores adversarial attacks on Graph Convolutional Transformer (GCT) models that utilize Electronic Health Record (EHR) data. As deep learning models become increasingly integral to healthcare, securing their robustness against adversarial threats is critical. This research assesses the susceptibility of GCT models to specific adversarial attacks, namely the Fast Gradient Sign Method (FGSM) and the Jacobian-based Saliency Map Attack (JSMA). It examines their effect on the model’s prediction of mortality and readmission. Through experiments conducted with the MIMIC-III and eICU datasets, the study finds that although the GCT model exhibits superior performance in processing EHR data under normal conditions, its accuracy drops when subjected to adversarial conditions—from an accuracy of 86% with test data to about 57% and an area under the curve (AUC) from 0.86 to 0.51. These findings averaged across both datasets and attack methods, underscore the urgent need for effective adversarial defense mechanisms in AI systems used in healthcare. This thesis contributes to the field by identifying vulnerabilities and suggesting various strategies to enhance the resilience of GCT models against adversarial manipulations.</p>

This research is describing a novel method to generate realism improved synthetic images while preserving annotation information and the eye gaze direction. Furthermore, it describes how the perceptual loss can be utilized while introducing basic features and techniques from adversarial networks for better results.