Pieter Gijsbers

Pieter Gijsbers

Auto-generated profile, based on publicly available information

Pieter Gijsbers

TUE & OpenML

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
30 January 2019 in Journal of Open Source Software
Verified
Journal Article
Open access
PDF file
PGPieter Gijsbers
Pieter Gijsbers
JVJoaquin Vanschoren
Joaquin Vanschoren
Successful machine learning applications hinge on a plethora of design decisions, which require extensive experience and relentless empirical evaluation. To train a successful model, one has to decide which algorithms to use, how to preprocess the data, and how to tune any hyperparameters that influence the final model. Automating this process of algorithm selection and hyperparameter optimization in the context of machine learning is often called AutoML (Automated Machine Learning).

9 June 2024
Verified
Conference Paper
Open access
MAMubashara Akhtar
Mubashara Akhtar
OBOmar Benjelloun
Omar Benjelloun
CCCostanza Conforti
Costanza Conforti
PGPieter Gijsbers
Pieter Gijsbers
JGJoan Giner-Miguelez
Joan Giner-Miguelez
CWCarole-Jean Wu
Carole-Jean Wu
Data is a critical resource for Machine Learning (ML), yet working with data remains a key friction point. This paper introduces Croissant, a metadata format for datasets that simplifies how data is used by ML tools and frameworks. Croissant makes datasets more discoverable, portable and interoperable, thereby addressing significant challenges in ML data management and responsible AI. Croissant is already supported by several popular dataset repositories, spanning hundreds of thousands of datasets, ready to be loaded into the most popular ML frameworks.

20 January 2018 in Information Visualization
Verified
Journal Article
Open access
MKMaximilian Konzack
Maximilian Konzack
PGPieter Gijsbers
Pieter Gijsbers
FTFerry Timmers
Ferry Timmers
ELEmiel van Loon
Emiel van Loon
MWMichel A Westenberg
Michel A Westenberg
KBKevin Buchin
Kevin Buchin
We present a visual analytics approach to explore and analyze movement data as collected by ecologists interested in understanding migration. Migration is an important and intriguing process in animal ecology, which may be better understood through the study of tracks for individuals in their environmental context. Our approach enables ecologists to explore the spatio-temporal characteristics of such tracks interactively. It identifies and aggregates stopovers depending on a scale at which the data is visualized. Statistics of stopover sites and links between them are shown on a zoomable geographic map which allows to interactively explore directed sequences of stopovers from an origin to a destination. In addition, the spatio-temporal properties of the trajectories are visualized by means of a density plot on a geographic map and a calendar view. To evaluate our visual analytics approach, we applied it on a data set of 75 migrating gulls that were tracked over a period of 3 years. The evaluation by an expert user confirms that our approach supports ecologists in their analysis workflow by helping to identifying interesting stopover locations, environmental conditions or (groups of) individuals with characteristic migratory behavior, and allows therefore to focus on visual data analysis.