Sections
Accueil UNamur > Agenda > Joint GRASCOMP and CIL Doctorat Training on Reproducible Research in Computer Science
événement

Joint GRASCOMP and CIL Doctorat Training on Reproducible Research in Computer Science

The Grascomp (GRAduate School in Computing Science) and CIL (Computational Intelligence and Learning) doctoral schools invite Ph.D. students, master's students about to start a Ph.D., and all interested researchers to a doctoral training event around reproducibility on Monday, July 6, in Louvain-la-Neuve.

Catégorie : conférence/cours/séminaire (spécialisé)
Date : 06/07/2026 09:00 - 06/07/2026 17:45
Lieu : UCLouvain, Louvain-la-Neuve, B059 auditorium of the Carnoy building

Research in computer science benefits from the ability to reproduce results and build upon past work. We have all been frustrated at least once when reading a research paper without access to code and data. An increasing number of computer science conferences in AI, cybersecurity, HCI, networks, and many more even require that artifacts be posted when submitting a paper. Artifact evaluation committees are also increasingly common, and allow research prototypes and datasets to be replayed and validated by peers.

In this doctoral training, we will cover the principles, tools, and good practices for reproducible research in computer science. We will cover examples and tools from different disciplines, technical ones, but also including those where humans are involved as subjects in the evaluation. We will also explore the existing and upcoming infrastructure that enables the deployment and evaluation of research prototypes.

The confirmed lineup of speakers is:

  • Prof. Valerio Schiavoni (UniNE, Switzerland) will share his experience and best practices for reproducible artifact generation, sharing, and evaluation in systems and security conferences.
  • Prof. Ana Lucic (University of Amsterdam) will cover reproducibility and open science for IA and machine learning.
  • Dr. Brecht Vermeulen (UGhent) will present the existing and upcoming infrastructures for experimental evaluation, and give an example of a reproducible experiment combining AI and systems aspects.
  • Prof. Jean Vanderdonckt (UCLouvain) will cover reproducibility in Human-Computer Interaction
  • Prof. Tom Barbette (UCLouvain) will cover best practices for reproducible performance and resource evaluation

Students interested in sharing their experience of producing and validating reproducible research are welcome to give a short address (within the available slots).

The event details are as follows:

Coffee breaks and lunch will be offered to all participants. Participants from outside UCLouvain are eligible for reimbursement of their travel (2nd-class train only) or may be provided with parking vouchers.

All Ph.D. students participating for the entire day will be provided with a certificate to claim ECTS credits with the local doctoral school.

Postdocs, faculty, and research engineers interested in infrastructure and reproducibility are also welcome to attend.

Registration is open until June 30 at https://forms.office.com/e/JXP0WSNMK5. Register ASAP, as we may have to close the form if the capacity is reached earlier than that.

Plus d'info : https://forms.office.com/e/JXP0WSNMK5
Télecharger : vCal