document

Causal Sensitivity Analysis: f-sensitivity through entropic value at risk computation

Havelka, Matej (author)

The field of causal inference provides a variety of estimators that can be used to find the effect of a treatment on an outcome based on observational data. However, many of these estimators require the unconfoundedness assumption, stating that all relevant confounders are observed within the data. This assumption is quite strict and many real...

master thesis 2024

document

Reliable Offine Policy Evaluation for Individualized Mechanical Ventilation

Volkers, Bas (author)

Individualizing mechanical ventilation treatment regimes remains a challenge in the intensive care unit (ICU). Reinforcement Learning (RL) offers the potential to improve patient outcomes and reduce mortality risk, by optimizing ventilation treatment regimes. We focus on the Offline RL setting, using Offline Policy Evaluation (OPE), specifically...

master thesis 2024

document

Reasoning about MDPs Abstractly: Bayesian Policy Search with Uncertain Prior Knowledge

Molhoek, Jord (author)

Many real-world problems fall in the category of sequential decision-making under uncertainty; Markov Decision Processes (MDPs) are a common method for modeling such problems. To solve an MDP, one could start from scratch or one could already have an idea of what good policies look like. Furthermore, there could be uncertainty in this idea. In...

master thesis 2024

document

Counterfactual Explanations of Learned Reward Functions

Wehner, Jan (author)

Learning rewards from humans is a promising approach to aligning AI with human values. However, methods are not able to consistently extract the correct reward functions from demonstrations or feedback. To allow humans to understand the limitations and misalignments of a learned reward function we adopt the technique of counterfactual...

master thesis 2023

document

Effectiveness of propensity score methods with density estimation in identifying overlap for causal inference

Tjong, Jonathan (author)

For causal inference, sufficient overlap is needed. It is possible to use propensity scores with the positivity assumption to ensure overlap is present. However, positivity is not enough to properly identify the region of overlap. For this, propensity scores need to be used in combination with density estimation. This project aims to evaluate...

bachelor thesis 2023

document

Using Nearest-Neighbors to Evaluate Overlap in Causal Inference

Vincenti, Jort (author)

To validate the results of a medical trial, there must be an overlap between the treatment and control groups. This implies the crucial need for good evaluation methods. This study, therefore, aimed to evaluate the overlap between causal classes using the Nearest Neighbours’ methods. Firstly, a case study was built around the common failures of...

bachelor thesis 2023

document

Combining Multi-Objective Planning with Reinforcement Learning to Solve Complex Tasks in Environments with Sparse Rewards

van Rijn, Cas (author)

Sequential decision-making problems are problems where the goal is to find a sequence of actions that complete a task in an environment. A particularly difficult type of sequential decision-making problem to solve is one in which the environment has sparse rewards, a large state space, and where the goal is to complete a complex task. In this...

master thesis 2023

document

Robustness of optimal randomized decision trees with dynamic programming

Götz, Valentijn (author)

Decision tree learning is widely done heuristically, but advances in the field of optimal decision trees have made them a more prominent subject of research. However, current methods for optimal decision trees tend to overlook the metric of robustness. Our research wants to find out whether the robustness of optimal decision trees can be...

bachelor thesis 2023

document

Optimal decision tree using dynamic programming: for the algorithm selection problem

Zeng, Henwei (author)

Several algorithms can often be used to solve a complex problem, such as the SAT problem or the graph coloring problem. Those algorithms differ in terms of speed based on the size or other features of the problem. Some algorithms perform much faster on a small size while others perform noticeably better on a larger instance. The optimization...

bachelor thesis 2023

document

Cost-sensitive optimal decision trees dealing with delays

Butzelaar, Sven (author)

Machine learning can be used to classify patients in a hospital. Here, the classifier has to minimize the cost of misclassifying the patient and minimize the costs of the tests. Unfortunately, obtaining features may be costly, e.g., taking blood tests or doing an x-ray scan. Furthermore, it is possible that acquiring those test results may take...

bachelor thesis 2023

document

DMQL: Deep Maximum Q-Learning: Combatting Relative Overgeneralisation in Deep Independent Learners using Optimism and Similarity

Dam, Erwin (author)

Various pathologies can occur when independent learners are used in cooperative Multi-Agent Reinforcement Learning. One such pathology is Relative Overgeneralisation, which manifests when a suboptimal Nash Equilibrium in the joint action space of a problem is preferred over an optimal Equilibrium. Approaches exist to combat relative...

master thesis 2022

document

Training human-AI agent in Overcooked

Vos, Jim (author)

Most cooperative games are tackled by creating a team of agents who are optimised for each other and the problem. Creating an agent who can play in a variety of teams without any foreknowledge of its partner is a different challenge. These AI systems could useful for human-AI interaction as different people bring a lot of variance into the...

bachelor thesis 2022

document

Cooperative multi-agent multi-target catching

Waij, Sander (author)

Multi-agent multi-target catching is the act of multiple agents trying to catch multiple moving targets. The exact algorithms created for such chases are generally targets that are actively trying to get as far as possible, while the agents try to prevent that. However, these algorithms can get much more efficient whenever targets also want to...

master thesis 2022

document

Frequency learning for structured CNN filters with Gaussian fractional derivatives

Saldanha, Nikhil (author)

A structured CNN filter basis allows incorporating priors about natural image statistics and thus require less training examples to learn, saving valuable annotation time. Here, we build on the Gaussian derivative CNN filter basis that learn both the orientation and scale of the filters. However, this Gaussian filter basis definition depends on...

master thesis 2021

document

Understanding Security and Privacy Characteristics ofHybrid-based IP-ICN Coexistence Architectures

Arora, Shruti (author)

Information-centric Networking (ICN) is the revolution of the internet due to its many benefit over the current internet infrastructure, namely caching, location-independent routing, and data-centric security. ICN ensures seamless data transfer due to its content-centric nature. Due to its content-centric nature, it is a must to ensure that...

bachelor thesis 2021

document

Privacy Preservation in a Blockchain-Based Healthcare System

Zagorac, Ivor (author)

The continuous generation of a large volume of health data from different sources has led to healthcare being a data-intensive domain. To achieve innovative advances in medical treatment procedures and to provide personalized healthcare services to the patients this data needs to be shared among different medical facilities. However, because...

bachelor thesis 2021

document

Security and privacy attacks and mitigations in Information-Centric Network

Rashad, Mohamed (author)

Information-Centric Networking (ICN) is a new approach for a more scalable and effective internet. ICN has many benefits, namely: ubiquitous caching, location-independent content routing and content-centric security. Despite the aforementioned benefits, the network paradigm is not ready to replace the current host-centric network as ICN is...

bachelor thesis 2021

document

Fine-grained Access Control for a Blockchain-based Healthcare System

Duveen, Sem (author)

Large volumes of medical data (MD) are continuously generated by the healthcare domain. When sharing these data, issues arise regarding privacy and security. To solve these issues, a permissioned blockchain (BC) can be used, but since blockchains do not have access control (AC) as a default feature, the integration of an access control system ...

bachelor thesis 2021

document

Integrating Consent Management Techniques into Blockchain-Based Medical Data Sharing

Jacobs, Ben (author)

The health industry is a data-intensive domain. Sharing medical data between medical facilities is necessary for providing good healthcare as well as for research in the healthcare domain. However, due to the sensitive and personal nature of health data, challenges arise when sharing this data. Consent management is a key aspect. The use of...

bachelor thesis 2021

document

Security and privacy in medical data sharing through blockchain

Guettache, Karim (author)

The sharing of medical data is becoming ever more important. More and more health-related data is being generated everyday and as will be shown later, its primary as well as its secondary usage brings many benefits to the healthcare system. However, medical data systems are not fail proof and are often the target of cyber-attacks, compromising...

bachelor thesis 2021