The Internet consists of many subnetworks, which are connected to each other. These subnetworks are the autonomous systems (ASes) that make up the Internet: each hosts a part of it. In order to successfully determine routes from one of these ASes to the other, the Border Gateway Protocol (BGP) is used. This protocol has several security flaws however, and exploitation of them has lead to parts of the Internet being temporarily unreachable. In order to combat these flaws, several security solutions have been developed already. However, none of these have been deployed on a wide scale yet. As such, this thesis focuses on the question: why not, and what can be done to protect BGP in the future? This thesis includes an analysis of the BGP threat landscape, to find which threats are most relevant, and to find out whether or not solutions have adapted to the threat landscape. It also includes a comparison of solutions on different practical security aspects. From this comparison, I found that no solution is able to prevent attacks if only one autonomous system deploys it. Due to this, I suggest to shift attention to detective security. This thesis also includes an analysis of some detective security schemes, to see which properties of these schemes can be used for a new scheme. This new scheme is designed to comply with a list of requirements, and it uses properties from three other schemes. Development of this scheme is left as future work. Altogether, this thesis should provide a new direction for the future of BGP security.

During a viral infection, we expel remnants of the virus. This makes it possible to conduct wastewater analysis which aid in the efforts to track the evolution of the current Covid-19 pandemic. It has been shown that by repurposing the kallisto algorithm, the abundance of SARS-CoV-2 variants in wastewater samples can be estimated. Since this is a novel method for this scope, its precision could probably be improved by adjusting certain aspects. In this work, I look at one of those aspects: sequencing particular genomic regions of the virus rather than the entire genome. I have indeed found that the regions that code for the spike (S) and nucleocapsid (N) regions and a section around the region coding for the non-structural protein 3 (nsp3) give particularly accurate results when sequenced on their own. In addition, in at least one case, combining two well-performing regions further improves accuracy at lower simulated abundances of variants. This suggests that sequencing depth is preferred over sequencing breath as long as the region being sequenced contains enough information to distinguish between variants. These findings are important as they can aid in the improvement of this method of variant quantification. Moreover, they can also help in improving other algorithms applied to the SARS-CoV-2 genome by highlighting the genomic sections containing the most differentiating information between variants.

Overcooked, an immersive multiplayer video game centered around cooperative cooking challenges, provides the roots for this research project. The study focuses on designing and evaluating a hand-authored controller in comparison to controllers implemented using various machine learning techniques, such as Population Based Training, in the context of a simplified version of the game. The main objective is to assess the cooperation between the hand-authored controller and a human controlled agent, with a particular emphasis on the coordination and minimizing errors during runs of fixed time length. A testing method has been designed in order to more accurately evaluate the performance. Through the implementation of a specialized controller utilizing techniques such as Behavior Trees and Decision Trees, the controller was able to successfully accomplish the task of delivering soups. An analysis was made to examine the performance of the hand-authored controller in comparison to the results obtained in other research papers that use the same simplified version of the game. After the results have been obtained, a clear difference was visible. The scripted AI performed better when paired with himself than one created with population based training that was paired with himself. However, the scripted AI was not better than a coupled planning algorithm that was paired with himself, but the computations were easier and faster for the scripted AI. When paired with a human, the overall performance decreases, but increases only for one specific map.

Cooperative AI is AI designed to cooperate with humans. One example of such an AI, made using planning algorithms, was studied in a paper from 2019 which used a simplified version of the video game Overcooked for evaluation. However, only limited evaluations were possible due to the long runtime and heuristic optimisations made. This paper will attempt to increase the amount of evaluations while removing an important optimisation limiting the functionality of the AI: the omission of counters. In the end it will find ways of reducing the runtime and increasing performance, which together allow for the addition of counters in a few instances. The final results with counters suggest there is more to gain upon removal of a select few heuristic optimisations and improvement of the human behavioural clone used to simulate a human.

The use of tangential vector fields and thus the need for designing them has steadily been increasing over the years. In this master thesis, a method is proposed and implemented that defines localized tangential vector fields on a mesh surface, which allows for the designing of vector fields on the triangulated surfaces of these meshes. Similarly, as with 2D radial basis functions for function values, it is possible to accurately approximate, compress and design vector fields using the localized tangential vector fields with an applied radial basis function scalar. The proposed system constructs multiple localized tangential vector fields on the surface and creates a basis from them that covers the entire surface. Other works have already suggested methods for the designing of tangential vector fields but most did not allow for both soft and hard constraints to be used when designing fields in real-time. One approach was able to design fields in real-time but had the bottleneck that it required a lot of precomputation time as it was using spectral processing techniques and a spectral transform to work in the spectral representation. Using the proposed localized tangential vector fields basis in this work also allows for the real-time modeling of tangential vector fields without to much precomputation time. However, it should be noted to fully optimize the proposed system several improvements on the current techniques used in this work have to be made to compete with the spectral representation. The proposed system is the first to our knowledge that uses localized tangential vector fields on meshes for the designing of tangential vector fields.

Procedural content generation (PCG) for architecture is widely used in a variety of digital media, most notably in games. However, such methods are often limited in their expressive range, and require considerable technical knowledge to create non-trivial architectural structures. We present a novel tile-based PCG approach for generating architecture, that proposes the use of architectural profiles, a declarative characterization of architectural typology, within a generic tile solving framework. An architectural profile consists of a set of tiles, representing atomic architectural building blocks, and a set of declarative constraints and rules, specifying which conditions a tile configuration has to satisfy to be valid. These conditions are translated into logic constraints, and used by a tile solver to control tile placement in a bottom-up manner. Eventually, each valid model output by the solver is a representative instance of its architectural profile. We describe an implementation of this approach with Answer Set Programming, using an off-the-shelf constraint solver for model generation. We performed an expressive range analysis, and concluded that our declarative method is quite controllable and can be steered over a broad range of architectural structures, regarding density and repetitiveness. Due to this expressive range and control, our tile-based method is very suitable for the customized development of urban environments for games. We also explore the adaptability of architectural profiles by application on pre-existing terrains.

The onset of delirium, a disturbance in the mental activities of a patient, can be potentially detected by understanding activities within an Intensive Care Unit (ICU) room. Such activities can be extracted by estimating human pose via a visual capture of the scene. This work uses a top-view depth camera in an ICU room to estimate pose of the non-patient stakeholders. The top-view leads to self-occlusions of body joints and thus poses a challenge for estimation of complete human pose. In addition, the presence of multiple persons in the room poses a secondary challenge, as detected body-joints need to be parsed into individual poses. To address these challenges, a 3D point cloud is extracted from the top-view depth image and passed through a 3D Convolutional Neural Network (CNN). This baseline method is capable of estimating both body-joints and body-parts to eventually output human pose for multiple persons. To improve the quality of output poses, the baseline method can benefit from additional spatial context since the problem of human pose estimation has a highly structured output. The proposed techniques either increase the receptive field, perform feature extraction at multiple scales or change the order of data processing. An increase in F1-score for the proposed methods highlights the importance of additional spatial context as a crucial tool to improve the performance of pose estimation models.

Accurate segmentation of anatomical structures and abnormalities in medical images is crucial, but manual segmentation is time-consuming and automated approaches lack clinical accuracy. In recent years, active learning approaches that aim to combine automatic segmentation with manual input have gained attention in the field, aiming to reduce the annotation effort required for training segmentation models. Batch query retrieval is a key component of active learning as it is a technique that allows for the simultaneous selection of multiple regions/points for annotation. This study investigates the effectiveness of discrete batch query retrieval methods compared to the traditional approach using gradient descent in the context of 3D medical image segmentation. Our experiments show that the active learning paradigm with batch query retrieval provides a few advantages over the gradient descent approach. Additionally, we analyze the impact of discretizing the query retrieval strategy on system performance. Our findings suggest that discretization can lead to slight performance degradation in terms of segmentation quality but offer computational advantages and faster convergence. We also discuss open issues, such as the interpretability of active learning methods, and recommend further research on combining active learning with other segmentation techniques. Overall, our study contributes to the understanding of active learning in medical image segmentation and provides insights for developing more efficient and accurate interactive segmentation models.

In ad-hoc cooperative environments, the usage of artificial intelligence to take supportive roles and work in collaboration with humans has proven to be of great benefit. The objective of this research is to evaluate the use of population-based training for reinforcement learning agents in a simplified version of the multiplayer game - Overcooked. The method used to answer that question involves evaluating the performance of the agents when paired with a human proxy and their learning curves on different layouts. Based on the employed method, it was concluded that both PBT and other self-play agents display notable underperformance when compared to human proxies and agents trained using human data. Moreover, while the inclusion of mutated agents enhanced sample efficiency in layouts with minimal collision risks, its effect on the final performance of PBT in those layouts was negligible. However, this approach managed to improve performance in layouts where collisions were the primary limiting factor.

Although automated segmentation of 3D medical images produce near-ideal results, they encounter limitations and occasional errors, necessitating manual intervention for error correction. Recent studies introduce an active learning pipeline as an efficient solution for this, requiring user corrections only on some of the most uncertain parts of the automatically segmented image. It does so by combining different uncertainty fields, which are various ways to quantify possible errors. We investigate into its individual uncertainty fields and their combination scheme in attempt to validate its methods. Additionally, we replace its methods for estimating uncertainty with another common way to do so, called the ensemble method, to test possible improvements at uncertainty estimation. Results of this research validates the combination method of the active learning pipeline, and shows weak advantages but strong disadvantages of the ensemble method when compared to the combined method of the active learning pipeline.

Current interconnected society provides us with numerous devices communicating with one another. Exchange of data thus become an integral part in our live. Data become valuable commodity in today's setting because of their usage by individual and other interested parties. Several parties may be interested in computing a function over their data while still want to keep the information on their own data private. Prior research on computing function in privacy preserving way in the domain of smart power-grid, e-metering system, wireless sensor network, and smart phone sensing generally focus on their own application and assume a total control and the static structure of the network. Moreover, a new paradigm in the field of decentralized power-grid requires privacy preserving solution to be applicable without existence of central authority. We propose two privacy preserving data aggregation protocols in peer-to-peer network scenario where there is such central authority involved. The first protocol utilizes additive homomorphism properties of Pailier scheme and the second protocol utilizes secret sharing. Both of the protocol achieve privacy-preserving requirement of some nodes in the network, as opposed to all nodes, that are included in the aggregation set by a hop count parameter from the initiating node. This way, both of the protocols require no information of overall network structure and privacy-preserving data aggregation is achieved by being able to communicate with direct neighbors of each node in the network only.

Convolutional Neural Networks (CNNs) are achieving state of the art performance in computer vision. One downside of CNNs is their computational complexity. One way to make CNNs more computational efficient is by implementing their convolutions in the frequency domain, using Fast Fourier Transforms (FFTs). This has as a consequence that most computational time in modern CNNs is spent in those FFTs. If all components of the CNN algorithm could be implemented in the frequency domain, it would no longer be necessary to go back at all to the spatial domain. Most components of the CNN do have alternatives that can be implemented in the frequency domain, however one crucial component doesn’t: the activation function. This thesis is the first published study into activation functions in the frequency domain. In this thesis several potential candidates for an activation function that works directly in the frequency domain are studied. Furthermore, some theoretical contributions on this subject are made.

This work investigates how prior knowledge from physics-based reflection models can be used to improve the performance of semantic segmentation models under an illumination-based domain shift. We implement various color invariants as a preprocessing step and find that CNNs trained on these color invariants get stuck in worse local minima compared to RGB inputs, but can achieve comparable or even superior performance when applying knowledge transfer from RGB. We also find Batch Normalization to severely affect the performance of neural networks under an illumination-based domain shift and demonstrate that Instance Normalization offers a simple remedy to this issue. Additionally, we investigate different fusion models for combining color invariants with RGB. Using a combination of these methods we achieve a 14.5% performance increase on nighttime semantic segmentation without any additional training data.

Urban change detection involves identifying and analyzing alterations in urban landscapes over time. This process is crucial for urban planning, environmental monitoring, and disaster management, as it provides insights into urban growth, land use changes, and human impact on the environment. This study focuses on Recurrent Neural Networks (RNNs) due to their ability to capture temporal dependencies, making them suitable for analyzing changes over time. In this research, an RNN model, specifically the SiamCRNN, was trained and evaluated on multiple datasets representing different urban scenarios. The model performed well in detecting urban changes, especially in datasets with higher spatial resolutions, but faced challenges with datasets characterized by high spectral range and complex urban structures. These findings underscore the importance of spatial resolution in influencing RNN model effectiveness for urban change detection tasks.

Graph Neural Networks (GNNs) have emerged as a powerful tool for learning from relational data. The real-world graphs such models are trained on are susceptible to changes in their topology. A growing body of work in the field of GNNs' stability to topology perturbations is trying to characterise how models respond to those changes, providing valuable insights that have enhanced the robustness of GNNs to adversarial attacks. The past work in this field, however, has only approached stability analysis using spectral graph theory, which is not applicable to all kinds of GNN models. In this paper, we aim to extend the past work in GNN stability by proposing an algorithm for analysing the stability of GNNs with model-agnostic GNN explainability tools instead of the mathematical framework of spectral graph theory. We demonstrate that the outputs of explainability tools can encode useful insights into the stability of GNNs and present a case study on using those insights to analyse node removal, edge removal, and edge weight perturbations.

Graph Neural Networks (GNNs) have emerged as a powerful tool for learning from relational data. The real-world graphs such models are trained on are susceptible to changes in their topology. A growing body of work in the field of GNNs' stability to topology perturbations is trying to characterise how models respond to those changes, providing valuable insights that have enhanced the robustness of GNNs to adversarial attacks. The past work in this field, however, has only approached stability analysis using spectral graph theory, which is not applicable to all kinds of GNN models. In this paper, we aim to extend the past work in GNN stability by proposing an algorithm for analysing the stability of GNNs with model-agnostic GNN explainability tools instead of the mathematical framework of spectral graph theory. We demonstrate that the outputs of explainability tools can encode useful insights into the stability of GNNs and present a case study on using those insights to analyse node removal, edge removal, and edge weight perturbations.

GNNs are a powerful tool for learning tasks on data with a graph structure. However, the topology of the graph in which GNNs are trained is often subject to change due to random, external perturbations. This research investigates the relationship between 5 topological properties of graphs (assortativity, density, edge connectivity, closeness centrality, diameter) and how stable GNNs trained on graphs with different topological properties are against different perturbations. The analysis is conducted by first synthetically generating graphs with different topological properties and training a GNN using the generated graphs. The synthetic graphs are then perturbed, and the relative change in the GNNs' output is measured. These results are further supported by conducting the same process on three popular GNN datasets: Cora, CiteSeer and PubMed citations. Finally, relationships between the graph properties under investigation and GNN stability are inferred using the results obtained from both synthetic and real-world datasets.

GNNs are a powerful tool for learning tasks on data with a graph structure. However, the topology of the graph in which GNNs are trained is often subject to change due to random, external perturbations. This research investigates the relationship between 5 topological properties of graphs (assortativity, density, edge connectivity, closeness centrality, diameter) and how stable GNNs trained on graphs with different topological properties are against different perturbations. The analysis is conducted by first synthetically generating graphs with different topological properties and training a GNN using the generated graphs. The synthetic graphs are then perturbed, and the relative change in the GNNs' output is measured. These results are further supported by conducting the same process on three popular GNN datasets: Cora, CiteSeer and PubMed citations. Finally, relationships between the graph properties under investigation and GNN stability are inferred using the results obtained from both synthetic and real-world datasets.