As the digital landscape evolves, cybersecurity has become an indispensable focus of IT systems. Its ever-escalating challenges have amplified the importance of digital forensics, particularly in the analysis of heap dumps from main memory. In this context, the Secure Shell protocol (SSH) designed for encrypted communications, serves as both a safeguard and a potential veil for malicious activities. This research project focuses on predicting SSH keys in OpenSSH memory dumps, aiming to enhance protective measures against illicit access and enable the development of advanced security frameworks or tools like honeypots. This Masterarbeit is situated within the broader SmartVMI project, and seeks to build upon existing research on key prediction in OpenSSH heap dumps. Utilizing machine learning (ML) and deep learning models, the study aims to refine features for embedding techniques and explore innovative methods for effective key detection based on recent advancements in Knowledge Graph and ML. The objective is to accurately predict the presence and location of SSH keys within memory dumps. This work builds upon, and aims to enhance, the foundations laid by SSHkex and SmartKex, enriching both the methodology and the results of the original research while exploring the untapped potential of newly proposed approaches. The current thesis dives into memory graph modelization from raw binary heap dump files. Each memory graph can support a range of embeddings that can be used directly for model training, through the use of classic ML models and graph neural network. It offers an in-depth discussion on the current state-of-the-art in key prediction for OpenSSH memory dumps, research questions, experimental setups, programs development, results as well as discussing potential future directions.

We propose a robust and reliable evaluation metric for generative models called Topological Precision and Recall (TopP&R, pronounced "topper"), which systematically estimates supports by retaining only topologically and statistically significant features with a certain level of confidence. Existing metrics, such as Inception Score (IS), Fréchet Inception Distance (FID), and various Precision and Recall (P&R) variants, rely heavily on support estimates derived from sample features. However, the reliability of these estimates has been overlooked, even though the quality of the evaluation hinges entirely on their accuracy. In this paper, we demonstrate that current methods not only fail to accurately assess sample quality when support estimation is unreliable, but also yield inconsistent results. In contrast, TopP&R reliably evaluates the sample quality and ensures statistical consistency in its results. Our theoretical and experimental findings reveal that TopP&R provides a robust evaluation, accurately capturing the true trend of change in samples, even in the presence of outliers and non-independent and identically distributed (Non-IID) perturbations where other methods result in inaccurate support estimations. To our knowledge, TopP&R is the first evaluation metric specifically focused on the robust estimation of supports, offering statistical consistency under noise conditions.

The advancement of Large Language Models (LLMs) has led to increasing concerns about the misuse of AI-generated text, and watermarking for LLM-generated text has emerged as a potential solution. However, it is challenging to generate high-quality watermarked text while maintaining strong security, robustness, and the ability to detect watermarks without prior knowledge of the prompt or model. This paper proposes an adaptive watermarking strategy to address this problem. To improve the text quality and maintain robustness, we adaptively add watermarking to token distributions with high entropy measured using an auxiliary model and keep the low entropy token distributions untouched. For the sake of security and to further minimize the watermark's impact on text quality, instead of using a fixed green/red list generated from a random secret key, which can be vulnerable to decryption and forgery, we adaptively scale up the output logits in proportion based on the semantic embedding of previously generated text using a well designed semantic mapping model. Our experiments involving various LLMs demonstrate that our approach achieves comparable robustness performance to existing watermark methods. Additionally, the text generated by our method has perplexity comparable to that of \emph{un-watermarked} LLMs while maintaining security even under various attacks.

Ensuring safe adoption of AI tools in healthcare hinges on access to sufficient data for training, testing and validation. In response to privacy concerns and regulatory requirements, using synthetic data has been suggested. Synthetic data is created by training a generator on real data to produce a dataset with similar statistical properties. Competing metrics with differing taxonomies for quality evaluation have been suggested, resulting in a complex landscape. Optimising quality entails balancing considerations that make the data fit for use, yet relevant dimensions are left out of existing frameworks. We performed a comprehensive literature review on the use of quality evaluation metrics on SD within the scope of tabular healthcare data and SD made using deep generative methods. Based on this and the collective team experiences, we developed a conceptual framework for quality assurance. The applicability was benchmarked against a practical case from the Dutch National Cancer Registry. We present a conceptual framework for quality assurance of SD for AI applications in healthcare that aligns diverging taxonomies, expands on common quality dimensions to include the dimensions of Fairness and Carbon footprint, and proposes stages necessary to support real-life applications. Building trust in synthetic data by increasing transparency and reducing the safety risk will accelerate the development and uptake of trustworthy AI tools for the benefit of patients. Despite the growing emphasis on algorithmic fairness and carbon footprint, these metrics were scarce in the literature review. The overwhelming focus was on statistical similarity using distance metrics while sequential logic detection was scarce. A consensus-backed framework that includes all relevant quality dimensions can provide assurance for safe and responsible real-life applications of SD.

Detecting discontinuity interfaces of discontinuous functions is a challenging task with significant implications across various scientific and engineering applications. Identifying these interfaces is particularly critical for functions with a high-dimensional domain, as their discontinuities can significantly influence the behavior of numerical methods and simulations; for example, within the realm of uncertainty quantification, where the smoothness of the target function plays a fundamental role in the use of stochastic collocation methods. Specifically, the knowledge of discontinuity interfaces enables the partitioning of the function domain into regions of smoothness, a crucial factor in improving the performance of numerical methods (e.g., see [17]). Other examples of discontinuity detection applications include signal processing, nonlinear partial differential equation (PDE) simulations, investigations of phase transitions in physical systems [14], and change-point analyses in geology or biology, to name a few [30]. The central objective of most discontinuity detection methods is to identify the position of discontinuities in the function domain using function evaluations on sets of points. Over the last few decades, progresses has been made in discontinuity detection, leading to the development of various algorithms. Notable works, such as [3, 2, 16, 35], have introduced significant contributions in this field. In particular, [3] introduced a polynomial annihilation edge detection method designed for piece-wise smooth functions with low-dimensional domains (n 2). This method identifies discontinuous interfaces by reconstructing jump functions based on a set of function evaluations.

Tissue segmentation is a routine preprocessing step to reduce the computational cost of whole slide image (WSI) analysis by excluding background regions. Traditional image processing techniques are commonly used for tissue segmentation, but often require manual adjustments to parameter values for atypical cases, fail to exclude all slide and scanning artifacts from the background, and are unable to segment adipose tissue. Pen marking artifacts in particular can be a potential source of bias for subsequent analyses if not removed. In addition, several applications require the separation of individual cross-sections, which can be challenging due to tissue fragmentation and adjacent positioning. To address these problems, we develop a convolutional neural network for tissue and pen marking segmentation using a dataset of 200 H&E stained WSIs. For separating tissue cross-sections, we propose a novel post-processing method based on clustering predicted centroid locations of the cross-sections in a 2D histogram. On an independent test set, the model achieved a mean Dice score of 0.981$\pm$0.033 for tissue segmentation and a mean Dice score of 0.912$\pm$0.090 for pen marking segmentation. The mean absolute difference between the number of annotated and separated cross-sections was 0.075$\pm$0.350. Our results demonstrate that the proposed model can accurately segment H&E stained tissue cross-sections and pen markings in WSIs while being robust to many common slide and scanning artifacts. The model with trained model parameters and post-processing method are made publicly available as a Python package called SlideSegmenter.

Artificial Intelligence (AI) finds widespread applications across various domains, sparking concerns about fairness in its deployment. While fairness in AI remains a central concern, the prevailing discourse often emphasizes outcome-based metrics without a nuanced consideration of the differential impacts within subgroups. Bias mitigation techniques do not only affect the ranking of pairs of instances across sensitive groups, but often also significantly affect the ranking of instances within these groups. Such changes are hard to explain and raise concerns regarding the validity of the intervention. Unfortunately, these effects largely remain under the radar in the accuracy-fairness evaluation framework that is usually applied. This paper challenges the prevailing metrics for assessing bias mitigation techniques, arguing that they do not take into account the changes within-groups and that the resulting prediction labels fall short of reflecting real-world scenarios. We propose a paradigm shift: initially, we should focus on generating the most precise ranking for each subgroup. Following this, individuals should be chosen from these rankings to meet both fairness standards and practical considerations.

Background and purpose: The unanticipated detection by magnetic resonance imaging (MRI) in the brain of asymptomatic subjects of white matter lesions suggestive of multiple sclerosis (MS) has been named radiologically isolated syndrome (RIS). As the difference between early MS [i.e. clinically isolated syndrome (CIS)] and RIS is the occurrence of a clinical event, it is logical to improve detection of the subclinical form without interfering with MRI as there are radiological diagnostic criteria for that. Our objective was to use machine-learning classification methods to identify morphometric measures that help to discriminate patients with RIS from those with CIS. Methods: We used a multimodal 3-T MRI approach by combining MRI biomarkers (cortical thickness, cortical and subcortical grey matter volume, and white matter integrity) of a cohort of 17 patients with RIS and 17 patients with CIS for single-subject level classification. Results: The best proposed models to predict the diagnosis of CIS and RIS were based on the Naive Bayes, Bagging and Multilayer Perceptron classifiers using only three features: the left rostral middle frontal gyrus volume and the fractional anisotropy values in the right amygdala and right lingual gyrus. The Naive Bayes obtained the highest accuracy [overall classification, 0.765; area under the receiver operating characteristic (AUROC), 0.782]. Conclusions: A machine-learning approach applied to multimodal MRI data may differentiate between the earliest clinical expressions of MS (CIS and RIS) with an accuracy of 78%. Keywords: Bagging; Multilayer Perceptron; Naive Bayes classifier; clinically isolated syndrome; diffusion tensor imaging; machine-learning; magnetic resonance imaging; multiple sclerosis; radiologically isolated syndrome.

Large language models have catalyzed an unprecedented wave in code generation. While achieving significant advances, they blur the distinctions between machine-and human-authored source code, causing integrity and authenticity issues of software artifacts. Previous methods such as DetectGPT have proven effective in discerning machine-generated texts, but they do not identify and harness the unique patterns of machine-generated code. Thus, its applicability falters when applied to code. In this paper, we carefully study the specific patterns that characterize machine and human-authored code. Through a rigorous analysis of code attributes such as length, lexical diversity, and naturalness, we expose unique pat-terns inherent to each source. We particularly notice that the structural segmentation of code is a critical factor in identifying its provenance. Based on our findings, we propose a novel machine-generated code detection method called DetectCodeGPT, which improves DetectGPT by capturing the distinct structural patterns of code. Diverging from conventional techniques that depend on external LLMs for perturbations, DetectCodeGPT perturbs the code corpus by strategically inserting spaces and newlines, ensuring both efficacy and efficiency. Experiment results show that our approach significantly outperforms state-of-the-art techniques in detecting machine-generated code.

Recent releases of pre-trained Large Language Models (LLMs) have gained considerable traction, yet research on fine-tuning and employing domain-specific LLMs remains scarce. This study investigates approaches for fine-tuning and leveraging domain-specific LLMs, highlighting trends in LLMs, foundational models, and methods for domain-specific pre-training. Focusing on the financial sector, it details dataset selection, preprocessing, model choice, and considerations crucial for LLM fine-tuning in finance. Addressing the unique characteristics of financial data, the study explores the construction of domain-specific vocabularies and considerations for security and regulatory compliance. In the practical application of LLM fine-tuning, the study outlines the procedure and implementation for generating domain-specific LLMs in finance. Various financial cases, including stock price prediction, sentiment analysis of financial news, automated document processing, research, information extraction, and customer service enhancement, are exemplified. The study explores the potential of LLMs in the financial domain, identifies limitations, and proposes directions for improvement, contributing valuable insights for future research. Ultimately, it advances natural language processing technology in business, suggesting proactive LLM utilization in financial services across industries.