document

BaSAL: Class Balanced Warm Start Active Learning for LiDAR Semantic Segmentation

Wei, Wei (author)

Active learning has been proposed as a solution to mitigate the expensive and time-consuming process of annotating large-scale autonomous driving datasets. The process typically involves a model initialization phase, followed by multiple iterations aiming at selecting the most informative data based on the initial model. However, we find two...

master thesis 2023

document

On Safety in Machine Learning

Viering, T.J. (author)

This dissertation focuses on safety in machine learning. Our adopted safety notion is related to robustness of learning algorithms. Related to this concept, we touch upon three topics: explainability, active learning and learning curves.<br/><br/>Complex models can often achieve better performance compared to simpler ones. Such larger models are...

doctoral thesis 2023

document

Knowledge- and ambiguity-aware robot learning from corrective and evaluative feedback

Celemin, Carlos (author), Kober, J. (author)

In order to deploy robots that could be adapted by non-expert users, interactive imitation learning (IIL) methods must be flexible regarding the interaction preferences of the teacher and avoid assumptions of perfect teachers (oracles), while considering they make mistakes influenced by diverse human factors. In this work, we propose an IIL...

journal article 2023

document

MORAL: Aligning AI with Human Norms through Multi-Objective Reinforced Active Learning

Peschl, M. (author), Zgonnikov, A. (author), Oliehoek, F.A. (author), Cavalcante Siebert, L. (author)

Inferring reward functions from demonstrations and pairwise preferences are auspicious approaches for aligning Reinforcement Learning (RL) agents with human intentions. However, state-of-the art methods typically focus on learning a single reward model, thus rendering it difficult to trade off different reward functions from multiple experts. We...

conference paper 2022

document

Showcase of Active Learning and Teaching Practices in Spatial Data Infrastructure (SDI) Education

Welle Donker, F.M. (author), van Loenen, B. (author), Kessler, Carsten (author), Küppers, Natalie (author), Panek, Mark (author), Mansourian, Ali (author), Zhao, Pengxiang (author), Vancauwenberghe, Glenn (author), Tomić, Hrvoje (author), Kević, Karlo (author)

The new concept of Open Spatial Data Infrastructures (Open SDIs) has emerged from an increased interest in open data initiatives together with national and international directives, such as the EU Open Data Directive (Directive (EU) 2019/1024), and the large investment of European public authorities in developing SDIs for sharing spatial data...

conference paper 2022

document

Gender, prior knowledge, and the impact of a flipped linear algebra course for engineers over multiple years

Hardebolle, Cécile (author), Verma, H. (author), Tormey, Roland (author), Deparis, Simone (author)

Background: Research shows that active pedagogies could play an important role in achieving more equitable outcomes for diverse groups of students in Science, Technology, Engineering, and Mathematics (STEM). Although flipped classes are a popular active methodology, there is a lack of high-quality studies assessing their impact in...

journal article 2022

document

To Actively Initialize Active Learning

Yang, Yazhou (author), Loog, M. (author)

Though much effort has been spent on designing new active learning algorithms, little attention has been paid to the initialization problem of active learning, i.e., how to find a set of labeled samples which contains at least one instance per category. This work identifies the initialization of active learning as a separate and novel...

journal article 2022

document

Does Personalization Help? Predicting How Social Situations Affect Personal Values

Kola, I. (author), Isufaj, R. (author), Jonker, C.M. (author)

Personal values represent what people find important in their lives, and are key drivers of human behavior. For this reason, support agents should provide help that is aligned with the personal values of the users. To do this, the support agent not only should know the value preferences of the user, but also how different situations in the...

conference paper 2022

document

Open Spatial Data Infrastructure Active Learning and Teaching Methods in Practice

Welle Donker, F.M. (author), van Loenen, B. (author), Poslončec Petrić, Vesna (author)

This paper describes active learning and teaching methods that can be used for teaching students new and complex concepts. Although active learning and teaching methods have been in use in the past in classrooms, the COVID-19 pandemic created new opportunities for adapting existing methods and developing methods suitable for online teaching. Not...

abstract 2022

document

Aligning AI with Human Norms: Multi-Objective Deep Reinforcement Learning with Active Preference Elicitation

Peschl, Markus (author)

The field of deep reinforcement learning has seen major successes recently, achieving superhuman performance in discrete games such as Go and the Atari domain, as well as astounding results in continuous robot locomotion tasks. However, the correct specification of human intentions in a reward function is highly challenging, which is why state...

master thesis 2021

document

Multi-AL: Robust Active learning for Multi-label Classifier

Basting, Mark (author)

Multi-label learning is becoming more and moreimportant as real-world data often contains multi-ple labels. The dataset used for learning such aclassifier is of great importance. Acquiring a cor-rectly labelled dataset is however a difficult task.Active learning is a method which can, given anoisy dataset, identify important instances for...

bachelor thesis 2021

document

Robust Multi-label Active Learning for Missing Labels

Rozen, Jonathan (author)

Multi-label classification has gained a lot of attraction in the field of computer vision over the past couple of years. Here, each instance belongs to multiple class labels simultaneously. There are numerous methods for Multi-label classification, however all of them make the assumption that either the training images are completely labelled or...

bachelor thesis 2021

document

Exploiting structure in counterexamples to speed up equivalence checking in the minimally adequate teacher framework: Active Learning

Catshoek, Tom (author)

Active state machine learning algorithms are a class of algorithms that allow us to infer state machines representing certain systems. These algorithms interact with a system and build a hypothesis of what the state machine describing that system looks like according to the behavior they observed. Once the algorithm arrives at a hypothesis, it...

master thesis 2021

document

Showcase of Existing Active Teaching Practices

Mansourian, Ali (author), Zhao, Pengxiang (author), Kessler, Carsten (author), Küppers, Natalie (author), Vancauwenberghe, Glenn (author), Lacroix, Lisa (author), Welle Donker, F.M. (author), van Loenen, B. (author), Tomić, Hrvoje (author), Kević, Karlo (author)

This report presents showcases of active teaching and learning in spatial data infrastructure education in the SPIDER partner universities. It includes detailed descriptions of the practices that have been implemented, as well as the results of the evaluation of the practices from an active teaching learning perspective.

report 2021

document

Methodology on Active Teaching and Learning on Open SDI

Pitz, Natalie (author), Schulze Althoff, Jan (author), Welle Donker, F.M. (author), van Loenen, B. (author), Vancauwenberghe, G. (author), Mansourian, Ali (author), Zhao, Pengxiang (author), Kević, Karlo (author), Tomić, Hrvoje (author)

This report looks at different methods of active teaching and learning and the application of these methods at the partner universities of the SPIDER project. Different methods of on-campus and online teaching are presented and reports on experiences in their application at the partner universities are discussed. In combination with the results...

report 2021

document

On-the-fly construction of surrogate constitutive models for concurrent multiscale mechanical analysis through probabilistic machine learning

Rocha, I.B.C.M. (author), Kerfriden, P. (author), van der Meer, F.P. (author)

Concurrent multiscale finite element analysis (FE²) is a powerful approach for high-fidelity modeling of materials for which a suitable macroscopic constitutive model is not available. However, the extreme computational effort associated with computing a nested micromodel at every macroscopic integration point makes FE²...

journal article 2021

document

A review and experimental analysis of active learning over crowdsourced data

Sayin, Burcu (author), Krivosheev, Evgeny (author), Yang, J. (author), Passerini, Andrea (author), Casati, Fabio (author)

Training data creation is increasingly a key bottleneck for developing machine learning, especially for deep learning systems. Active learning provides a cost-effective means for creating training data by selecting the most informative instances for labeling. Labels in real applications are often collected from crowdsourcing, which engages...

journal article 2021

document

Population-based Active Learning for Black-Box Regression

van Deursen, Max (author)

Many applications employ models to represent real-life environments efficiently. To allow these models to be realistic it is commonly fitted using a dataset containing labeled samples. When obtaining a label for a sample from the environment is expensive, it is key that the dataset contains only those samples that aid in providing a realistic...

master thesis 2020

document

Sampling settings in active learning for investigating inconsistency

Li, Mengze (author)

Active learning has the potential to reduce labeling costs in terms of time and money. In practical use, active learning works as an efficient data labeling strategy. Another point of view to look at active learning is to consider active learning as a learning problem, where the training data is queried by the active learner. Under this...

master thesis 2020

document

Discrete state-space active inference and unknown controlled state transitions

van Roessel, L. (author)

Active inference is a process theory arising from neuroscience which casts perception, action, planning and learning under one optimisation criterion: minimisation of free energy. Current literature on the implementation of discrete state-space active inference focuses on scalability, the comparison to reinforcement learning and its performance...

master thesis 2020