Binnen het brede netwerk van de Universiteit Leiden neemt het instituut deel aan onderzoek binnen de volgende profileringsgebieden:

The focus of this thesis is on the technical methods which help promote the movement towards Trustworthy AI, specifically within the Inspectorate of the Netherlands.

Transport inspectorates are looking for novel methods to identify dangerous behavior, ultimately to reduce risks associated to the movements of people and goods. We explore a data-driven approach to arrive at smart inspections of vehicles.

In the era of big data, extracting insights from complex datasets is a key challenge. This thesis demonstrates the superiority of graph-based methods over traditional clustering (e.g., k-means, DBSCAN) and outlier detection for analyzing high-dimensional and noisy data.

Dit project beoogt de ontwikkeling van geavanceerde data-analysemethoden voor monitoring en het vergroten van ons inzicht in de dynamiek van de biodiversiteit in natuurgebieden zoals de Oostvaardersplassen.

This thesis mainly focuses on multimodal understanding and Visual Question Answering (VQA) via deep learning methods. For technical contributions, this thesis first focuses on improving multimodal fusion schemes via multi-stage vision-language interactions.

Promotor: Prof.dr. H.A.G. Wijshoff

In this thesis, we have studied cell images from two types of cells, including pollen grains and the immune cells, neutrophils. These images are captured using a bright field microscope and a confocal microscope.

Promotor: F. Arbab, R. De Nicola, Co-Promotor: M. Tribastone

Uncertainty and noise are frequently-encountered obstacles in real-world applications of numerical optimization. The practice of optimization that deals with uncertainties and noise is commonly referred to as robust optimization.

In this work we describe three methods to improve the performance of Quantum Field Theory calculations.

Today, knowledge is the most crucial element to stimulate organizational competitiveness and economic development.

Automated Machine Learning (AutoML) is widely used to automatically build a suitable practical Machine Learning (ML) model for an arbitrary real-world problem, reducing the effort of practitioners in the ML development cycle for real-world applications. Optimization is a key part of a typical AutoML…

Reinforcement learning is a framework that enables agents to learn in a manner similar to humans, i.e. through trial and error. Ideally, we would like to train a generalist agent capable of performing multiple tasks and achieving various goals.

Humans share meaning through language. Over time, repeated interactions have shaped languages into forms that match our cognitive preferences, making them structured, expressive, easy to learn, and ultimately, meaningful.

Quantum computers are a proposed fundamentally new type of computer. They aim to perform some computations much faster than previously possible by exploiting phenomena at the quantum scale, called superposition and entanglement.

The database research community has made tremendous strides in developing powerful database engines that allow for efficient analytical query processing.

In this thesis, topics relating to the optimization of high-throughput pipelines used for imaging are discussed. In particular, different levels of implementation, i.e., conceptual, software, and hardware, are discussed and the thesis outlines how advances on each level need to be made to make gains…

De faculteit biedt een internationale academische omgeving met 36% van de studenten, promovendi en onderzoekers uit het buitenland.

Onderzoekers van LIACS werken aan het bedenken van de computers van morgen die de ruggengraat zullen vormen van de Cloud en Edge computing paradigma’s en ‘the Internet of Things’. In dit verband zijn we betrokken bij onderzoek en ontwikkeling van high performance computing systemen, embedded & real-time…

Promotor: J. N. Kok, Co-promotor: A. J. Knobbe

With the dramatic growth of the image data on the web, there is an increasing demand of the algorithms capable of understanding the visual information automatically.

For optimization problems, it is often unclear how to choose the most appropriate optimization algorithm. As such, rigorous benchmarking practices are critical to ensure we can gain as much insight into the strengths and weaknesses of these types of algorithms.

Quantum hardware comes with a different computing paradigm and new ways to tackle applications. Much effort has to be put into understanding how to leverage this technology to give real-world advantages in areas of interest for industries such as combinatorial optimization or machine learning.

Promotor: F. Arbab

Optimization tasks in practice have multifaceted challenges as they are often black box, subject to multiple equality and inequality constraints and expensive to evaluate.

This thesis applies data mining in commercial settings in the telecommunications industry.

Promotor: Prof.dr. J.N. Kok, Co-promotor: A.J. Knobbe

Anticipatie op de uitdagingen en voordelen die AI-ondersteunde beoordelingen met zich meebrengen voor de relatie tussen studenten en docenten, de menselijke, evenals sociaal-economische kosten en kansen die zij voor universiteiten creëren.

Promotor: J.N. Kok, P.E. Slagboom Co-promotor: J.F.J. Laros

Constraint multi-objective optimization with a limited budget for function evaluations is challenging. This thesis tackles this problem by proposing new optimization algorithms. These algorithms are applied on holistic ship design problems. This helps naval architects balance objectives like cost, efficiency,…

The increasing competition in the automotive industry requires the tailored, swift development of technologically sophisticated vehicles. Therefore, the computationally expensive state-of-the-art simulation technologies are combined with optimization algorithms. An example of a real-world optimization…

This thesis seeks to combine two different research topics; Multi-Objective Optimization and Complex Network Analysis.

Promotor: J.N. Kok, Co-promotor: C.M. Soares, A.J. Knobbe

SuperCode: Sustainability PER AI-driven CO-DEsign

Mining time series is a machine learning subfield that focuses on a particular data structure, where variables are measured over (short or long) periods of time.

Promotor: Prof.dr. E. Deprettere

Promotores: T.H.W. Bäck, Y. Tan, Co-promotor: M.T.M. Emmerich

Promotor: Prof.dr. B.R. Katzy, Prof.dr. H.J. van den Herik, Prof.dr. G.H. Baltes (University of Applied Sciences Konstanz)

This thesis is about augmented reality (AR). AR is commonly considered a technology that integrates virtual images into a user’s view of the real world.

Promotores: T.H.W. Bäck, A. Plaat, Co-promotor: F.J. Verbeek

By training with virtual opponents known as computer generated forces (CGFs), trainee fighter pilots can build the experience necessary for air combat operations, at a fraction of the cost of training with real aircraft.

Big Science, commonly defined as conventional science made big in three dimensions, namely organizations, machines, and politics, brings a plethora of different stakeholders together, often for a long period of time. This includes policymakers, scientists, (scientific) managers as well as local “host”…

This dissertation explores how Large Language Models (LLMs) can effectively and responsibly contribute to complex decision-making processes. By combining AI and human intelligence, Hybrid Intelligence (HI) emerges, allowing the strengths of both humans and machines to be utilized.

Promotores: Prof.dr. F.S. de Boer, Prof.dr. J.J.M.M. Rutten (Radboud Universiteit Nijmegen)

In this work, we attempt to answer the question:

In 2018, the number of mobile phone users will reach about 4.9 billion. Assuming an average of 5 photos taken per day using the built-in cameras would result in about 9 trillion photos annually.

Divide & Quantum (D&Q) biedt verschillende oplossingen om de kracht van quantumcomputers op korte termijn te benutten, en stelt volledige pipelines voor, van theoretisch onderzoek, via implementatie tot real-world case studies in verschillende disciplines, tot wetenschapscommunicatie naar een bredere…

Image segmentation is a fundamental task in computer vision, with applications ranging from medical diagnostics to archaeological research.

Natural history collections provide invaluable sources for researchers with different disciplinary backgrounds, aspiring to study the geographical distribution of flora and fauna across the globe as well as other evolutionary processes.