This dissertation introduces a comprehensive framework aimed at reshaping applied cybersecurity education to significantly ease the learning curve, at scale, through three synergistic innovations. These methods address the daunting educational barriers in cybersecurity, enabling learners at all levels to understand complex security concepts more easily. The first innovation, the PWN methodology, redefines the traditional Capture The Flag (CTF) model by offering a structured series of modularized, self-guided challenges. This approach helps simplify complex topics into manageable units, each building on the last, which allows students to progress at their own pace. Over five years and with over 400 systems security challenges developed, this method has effectively helped students evolve from beginners to masters of advanced security exploits. The second component is the DOJO platform, an open-source learning environment that uses containerization technology to provide a pre-configured, browser-based interface. This platform reduces the setup complexities associated with applied cybersecurity and has already given over 10,000 students immediate access to practical learning scenarios, from vulnerability discovery to advanced debugging, in a unified, user-friendly environment. Its seamless integration allows educators to quickly launch new challenges and resources, ensuring a continuous and dynamic educational experience. The third component, the SENSAI tutor, is an AI-driven tutoring system that leverages Large Language Models to offer personalized, intelligent support. Integrated with the PWN methodology and DOJO platform, SENSAI serves as an on-demand mentor, providing tailored advice and problem-solving assistance. It adapts to individual student needs, offering specific guidance and theoretical support to enhance understanding and retention of complex concepts. Together, these three components create a powerful, integrated educational strategy that not only equips students with vital cybersecurity skills but also deepens their understanding of digital vulnerabilities and the strategic thinking needed to mitigate them. This strategy prepares a new generation of cybersecurity professionals to navigate the ever-evolving threats of the digital world.

Having properly implemented instructions is key to computer architecture and the security of a computer. Without properly implemented instructions, there is a risk of security vulnerabilities such as privilege escalation. Current methods of checking specification mismatches are the various versions of the manual approach and the use of symbolic execution. These current methods can be time-consuming or have issues with scalability and efficiency. In this thesis, an approach is proposed to improve the current methods by employing the aid of machine-learning, specifically large-language models (LLMs), testing on RISC-V architecture. RISC-V architecture is proposed to test this method due to its simplistic nature and smaller instruction set compared to other architectures like x86. In this approach, Chat-GPT is proposed as the LLM of choice due to its rising popularity as well as its capability and power. The approach combines manual aspects and the aid of Chat-GPT to fully test how well Chat-GPT is at generating expressions and test cases to detect specification mismatches. The Chat-GPT generated test cases are evaluated on a RISC-V framework to see if the Chat-GPT generated test cases can be used in the future to detect specification mismatches as well as being used in more complicated architectures.

Learning longer-horizon tasks is challenging with techniques such as reinforcement learning and behavior cloning. Previous approaches have split these long tasks into shorter tasks that are easier to learn by using statistical change point detection methods. However, classical changepoint detection methods function only with low-dimensional robot trajectory data and not with high-dimensional inputs such as vision. In this thesis, I have split a long horizon tasks, represented by trajectories into short-horizon sub-tasks with the supervision of language. These shorter horizon tasks can be learned using conventional behavior cloning approaches. I found comparisons between the techniques from the video moment retrieval problem and changepoint detection in robot trajectory data consisting of high-dimensional data. The proposed moment retrieval-based approach shows a more than 30% improvement in mean average precision (mAP) for identifying trajectory sub-tasks with language guidance compared to that without language. Several ablations are performed to understand the effects of domain randomization, sample complexity, views, and sim-to-real transfer of this method. The data ablation shows that just with a 100 labeled trajectories a 42.01 mAP can be achieved, demonstrating the sample efficiency of using such an approach. Further, behavior cloning models trained on the segmented trajectories outperform a single model trained on the whole trajectory by up to 20%.

Efficiently solving global optimization problems remains a pervasive challenge across diverse domains, characterized by complex, high-dimensional search spaces with non-convexity and noise. Most of the approaches in the Bayesian optimization literature have highlighted the computational complexity involved when scaling to high dimensions. Non myopic approximations over a finite horizon has been adopted in recent years by modeling the problem as a partially observable Markov Decision Process (MDP). Another promising direction is the partitioning of the input domain into sub regions facilitating local modeling of the input space. This localized approach helps prioritize regions of interest, which is particularly crucial in high dimensions. However, very few literature exist which leverage agent based modeling of the problem domain along with the aforementioned methodologies. This work explores the synergistic integration of Bayesian Optimization and Reinforcement Learning by proposing a Multi Agent Rollout formulation of the global optimization problem. Multi Agent Bayesian Optimization (MABO) partitions the input domain among a finite set of agents enabling distributed modeling of the input space. In addition to selecting candidate samples from their respective sub regions, these agents also influence each other in partitioning the sub regions. Consequently, a portion of the function is optimized by these agents which prioritize candidate samples that don't undermine exploration in favor of a single step greedy exploitation. This work highlights the efficacy of the algorithm on a range of complex synthetic test functions.

Manipulator motion planning has conventionally been solved using sampling and optimization-based algorithms that are agnostic to embodiment and environment configurations. However, these algorithms plan on a fixed environment representation approximated using shape primitives, and hence struggle to find solutions for cluttered and dynamic environments. Furthermore, these algorithms fail to produce solutions for complex unstructured environments under real-time bounds. Neural Motion Planners (NMPs) are an appealing alternative to algorithmic approaches as they can leverage parallel computing for planning while incorporating arbitrary environmental constraints directly from raw sensor observations. Contemporary NMPs successfully transfer to different environment variations, however, fail to generalize across embodiments. This thesis proposes "AnyNMP'', a generalist motion planning policy for zero-shot transfer across different robotic manipulators and environments. The policy is conditioned on semantically segmented 3D pointcloud representation of the workspace thus enabling implicit sim2real transfer. In the proposed approach, templates are formulated for manipulator kinematics and ground truth motion plans are collected for over 3 million procedurally sampled robots in randomized environments. The planning pipeline consists of a state validation model for differentiable collision detection and a sampling based planner for motion generation. AnyNMP has been validated on 5 different commercially available manipulators and showcases successful cross-embodiment planning, achieving an 80% average success rate on baseline benchmarks.

Traffic monitoring plays a crucial role in urban planning, transportation management, and road safety initiatives. However, existing monitoring systems often struggle to balance the need for high-resolution data acquisition and resource efficiency. This study proposes an innovative approach leveraging neuromorphic sensor technology to enhance traffic monitoring efficiency while still exhibiting robust performance when exposed to difficult conditions. Neuromorphic cameras, also called event-based cameras, with their high temporal and dynamic range and minimal memory usage, have found applications in various fields. However, despite their potential, their use in static traffic monitoring is largely unexplored. This study introduces eTraM, the first-of-its-kind fully event-based traffic monitoring dataset, to address the gap in existing research. eTraM offers 10 hr of data from diverse traffic scenarios under varying lighting and weather conditions, providing a comprehensive overview of real-world situations. Providing 2M bounding box annotations, it covers eight distinct classes of traffic participants, ranging from vehicles to pedestrians and micro-mobility. eTraM's utility has been assessed using state-of-the-art methods, including RVT, RED, and YOLOv8. The quantitative evaluation of the ability of event-based models to generalize on nighttime and unseen scenes further substantiates the compelling potential of leveraging event cameras for traffic monitoring, opening new avenues for research and application.

Rapid advancements in artificial intelligence (AI) have revolutionized various do- mains, enabling the development of sophisticated models capable of solving complex problems. However, as AI systems increasingly participate in critical decision-making processes, concerns about their interpretability, robustness, and reliability have in- tensified. Interpretable AI models, such as the Concept-Centric Transformer (CCT), have emerged as promising solutions to enhance transparency in AI models. Yet, in- creasing model interpretability often requires enriching training data with concept ex- planations, escalating training costs. Therefore, intrinsically interpretable models like CCT must be designed to be data-efficient, generalizable—to accommodate smaller training sets—and robust against noise and adversarial attacks. Despite progress in interpretable AI, ensuring the robustness of these models remains a challenge.This thesis enhances the data efficiency and generalizability of the CCT model by integrating four techniques: Perturbation Random Masking (PRM), Attention Random Dropout (ARD), and the integration of manifold mixup and input mixup for memory broadcast. Comprehensive experiments on benchmark datasets such as CIFAR-100, CUB-200-2011, and ImageNet show that the enhanced CCT model achieves modest performance improvements over the original model when using a full training set. Furthermore, this performance gap increases as the training data volume decreases, particularly in few-shot learning scenarios. The enhanced CCT maintains high accuracy with limited data (even without explicitly training on ex- ample concept-level explanations), demonstrating its potential for real-world appli- cations where labeled data are scarce. These findings suggest that the enhancements enable more effective use of CCT in settings with data constraints. Ablation studies reveal that no single technique—PRM, ARD, or mixups—dominates in enhancing performance and data efficiency. Each contributes nearly equally, and their combined application yields the best results, indicating a synergistic effect that bolsters the model’s capabilities without any single method being predominant. The results of this research highlight the efficacy of the proposed enhancements in refining CCT models for greater performance, robustness, and data efficiency. By demonstrating improved performance and resilience, particularly in data-limited sce- narios, this thesis underscores the practical applicability of advanced AI systems in critical decision-making roles.

In the age of artificial intelligence, Machine Learning (ML) has become a pervasive force, impacting countless aspects of our lives. As ML’s influence expands, concerns about its reliability and trustworthiness have intensified, with security and robustness emerging as significant challenges. For instance, it has been demonstrated that slight perturbations to a stop sign can cause ML classifiers to misidentify it as a speed limit sign, raising concerns about whether ML algorithms are suitable for real-world deployments. To tackle these issues, Responsible Machine Learning (Responsible ML) has emerged with a clear mission: to develop secure and robust ML algorithms. This dissertation aims to develop Responsible Machine Learning algorithms under real-world constraints. Specifically, recognizing the role of adversarial attacks in exposing security vulnerabilities and robustifying the ML methods, it lays down the foundation of Responsible ML by outlining a novel taxonomy of adversarial attacks within real-world settings, categorizing them into black-box target-specific, and target-agnostic attacks. Subsequently, it proposes potent adversarial attacks in each category, aiming to obtain effectiveness and efficiency. Transcending conventional boundaries, it then introduces the notion of causality into Responsible ML (a.k.a., Causal Responsible ML), presenting the causal adversarial attack. This represents the first principled framework to explain the transferability of adversarial attacks to unknown models by identifying their common source of vulnerabilities, thereby exposing the pinnacle of threat and vulnerability: conducting successful attacks on any model with no prior knowledge. Finally, acknowledging the surge of Generative AI, this dissertation explores Responsible ML for Generative AI. It introduces a novel adversarial attack that unveils their adversarial vulnerabilities and devises a strong defense mechanism to bolster the models’ robustness against potential attacks.

Astronomy has a data de-noising problem. The quantity of data produced by astronomical instruments is immense, and a wide variety of noise is present in this data including artifacts. Many types of this noise are not easily filtered using traditional handwritten algorithms. Deep learning techniques present a potential solution to the identification and filtering of these more difficult types of noise. In this thesis, deep learning approaches to two astronomical data de-noising steps are attempted and evaluated. Pre-existing simulation tools are utilized to generate a high-quality training dataset for deep neural network models. These models are then tested on real-world data. One set of models masks diffraction spikes from bright stars in James Webb Space Telescope data. A second set of models identifies and masks regions of the sky that would interfere with sky surface brightness measurements. The results obtained indicate that many such astronomical data de-noising and analysis problems can use this approach of simulating a high-quality training dataset and then utilizing a deep learning model trained on that dataset.

With the exponential growth of multi-modal data in the field of computer vision, the ability to do inference effectively among multiple modalities—such as visual, textual, and auditory data—shows significant opportunities. The rapid development of cross-modal applications such as retrieval and association is primarily attributed to their ability to bridge the gap between different modalities of data. However, the current mainstream cross-modal methods always heavily rely on the availability of fully annotated paired data, presenting a significant challenge due to the scarcity of precisely matched datasets in real-world scenarios. In response to this bottleneck, several sophisticated deep learning algorithms are designed to substantially improve the inference capabilities across a broad spectrum of cross-modal applications. This dissertation introduces novel deep learning algorithms aimed at enhancing inference capabilities in cross-modal applications, which take four primary aspects. Firstly, it introduces the algorithm for image retrieval by learning hashing codes. This algorithm only utilizes the other modality data in weakly supervised tags format rather than the supervised label. Secondly, it designs a novel framework for learning the joint embeddings of images and texts for the cross-modal retrieval tasks. It efficiently learns the binary codes from the continuous CLIP feature space and can even deliver competitive performance compared with the results from non-hashing methods. Thirdly, it conducts a method to learn the fragment-level embeddings that capture fine-grained cross-modal association in images and texts. This method uses the fragment proposals in an unsupervised manner. Lastly, this dissertation also outlines the algorithm to enhance the mask-text association ability of pre-trained semantic segmentation models with zero examples provided. Extensive future plans to further improve this algorithm for semantic segmentation tasks will be discussed.