News
2024
- The paper “A Uniformly Bounded Correlation Function for Spatial Point Patterns” by Evgenia Martynova and Johannes Textor was accepted at KDD 2024.
- The paper “Population-Based Algorithms Built on Weighted Automata” by Gijs Schröder, Inge Wortel and Johannes Textor was accepted at PPSN 2024.
- Gijs Hendriksen, Djoerd Hiemstra co-edited the Proceedings of the first International Workshop on Open Web Search, published in the CEUR Workshop Series as Volume 3689. https://ceur-ws.org/Vol-3689/
- Faegheh Hasibi gave a presentation at the AI4Science meet up at the Faculty of Science about “Knowledge-Grounded Information Seeking with LLMs.”
- Heydar Soudani, Faegheh Hasibi, Roxana Petcu (UvA) and Evangelos Kanoulas (UvA) presented a tutorial at the Web Conference 2024 in Signapore about “Data Augmentation for Conversational AI.”
- Faegheh Hasibi was one the members of the Radboud team to visit ASML and gave a talk on the “usage of LLMs for building chatbots in domain-specific applications.”
- Paper has been accepted at ACL findings 2024 by Mohanna Hoveyda, Arjen P. de Vries, Maarten de Rijke, and Faegheh Hasibi : “Real World Conversational Entity Linking Requires More Than Zero-Shots.”
- Heydar Soudani, Roxana Petcu (UvA), Evangelos Kanoulas (UvA) and Faegheh Hasibi published a new pre-print: “A Survey on Recent Advances in Conversational Data Generation“
- The interdisciplinary project proposal 'Self Reflection Machines' with colleagues Marieke de Vries and Marene Dimmendaal (iCIS), Hanna Schraffenberger (iCIS, iHUB), Pim Haselager and Simon Fischer (AI, Donders) has been awarded with the ‘Radboud Young Academy Cross-Faculty Catapult Grant’. The project will focus on supporting patients in reflecting on a decision that they have to make regarding their treatment plan. Specifically, it would support their evaluation of how the treatment might impact their personal life and whether their decision matches personal beliefs, values and preferences.
- Harrie Oosterhuis, Emma Gerritse, Arjen de Vries and Djoerd Hiemstra visited two elementary schools to teach 10-12 year-old children about information retrieval and recommendation systems. These visits were organized by the Science Education Hub Radboud University (WKRU) as part of the Radboud Science Award prize that Harrie received in 2023.
- Harrie Oosterhuis will receive the Andreas Bonn medal in the domain of science (physics, chemistry, math and computer science) for his 2020 PhD thesis. This prize is awarded once every five years to a PhD thesis from the University of Amsterdam or the Vrije Universiteit Amsterdam.
- Three papers co-authored by DaS members were accepted at the ACM SIGIR Conference on Research and Development in Information Retrieval in Washington DC: 1) “Going Beyond Popularity and Positivity Bias: Correcting for Multifactorial Bias in Recommender Systems”, Jin Huang, Harrie Oosterhuis, Masoud Mansoury, Herke van Hoof and Maarten de Rijke. 2) “Estimating the Hessian Matrix of Ranking Objectives for Stochastic Learning to Rank with Gradient Boosted Trees”, Jingwei Kang, Maarten de Rijke and Harrie Oosterhuis. 3) “Doing Personal LAPS: LLM-Augmented Dialogue Construction for Personalized Multi-Session Conversational Search”, Shubham Chatterjee, Jeff Dalton (University of Edinburg), Andrew Ramsay (University of Glasgow), Hideaki Joko, Arjen P. de Vries, Faegheh Hasibi (Radboud University). This project is conducted through “Radboud- Glasgow collaboration fund.”
- Tom Heskes and Roel Bouman, in collaboration with Zaharah Bukhsh, have had their paper: “Unsupervised Anomaly Detection Algorithms on Real-world Data: How Many Do We Need?” published in the Journal of Machine Learning Research: https://www.jmlr.org/papers/volume25/23-0570/23-0570.pdf
- Gijs Hendriksen and Negin Ghasemi presented their work at the 46th European Conference on Information Retrieval (ECIR) in Glasgow.
- Arjen de Vries won the outstanding reviewer award at ECIR in Glasgow.
- Yuliya Shapovalova delivered a keynote at the IRP Symposium on the topic of ‘Interdisciplinary Approach to Machine Learning”.
- Tom Heskes was one of the speakers at the Innovate event on AI in a packed Lindenberg. https://www.innovate.community/event/innovate-ai-meetup-2024/
- Partnership event Alliander & Radboud Universiteit. The event took place on Thursday 29 February 2024 with contributions from Yuliya Shapovalova, Charlotte Cambier van Nooten, Roel Bouman, and Olivier Claessen. https://www.ru.nl/en/about-us/news/collaboration-radboud-and-alliander-extended
- Djoerd Hiemstra gave his inaugural lecture on Friday 1 March 2024 at 15:45h.: Zoekmachines: Samen en duurzaam vooruit. https://djoerdhiemstra.com/2024/search-engines-together-and-sustainably/
- Gijs Hendriksen and Djoerd Hiemstra co-organize the first International Workshop on Open Web Search (WOWS) co-locationed with the European Conference on Information Retrieval (ECIR) on 28 March in Glasgow: https://opensearchfoundation.org/wows2024/.
- On Wednesday 15th January, we had our first New Year’s Poster Session on our floor. During this poster session, we could update each other on our previous, ongoing, and future work. It was a great success as multiple colleagues came up with new ideas! We are looking forward to the next poster session.
- On Monday January 22nd, Feri Wijayanto successfully defended his PhD thesis titled: Automated Model Selection for Rasch and Mediation Analysis.
2023
- Harrie Oosterhuis was part of the organization of the 2nd edition of the CONSEQUENCES workshop at the 2023 ACM RecSys conference, speakers at the workshop included Prof. Micheal I. Jordan from Berkeley and dr. Minmin Chen from Google. https://sites.google.com/view/consequences2023/home
- Harrie Oosterhuis is one of three winners of this year's Radboud science awards for his research on search and recommendation: https://www.ru.nl/en/about-us/news/radboud-science-awards-2023-conscientious-language-searching-computers-and-smell
- Harrie Oosterhuis received a NWO Veni grant for his project titled: "Search Systems with a Bayesian Understanding of Their Users (BAYES-SEARCH)": https://www.ru.nl/en/research/research-news/nwo-veni-grants-for-research-into-frontline-workers-social-media-and-more
- The data science group had five paper presentations and a half-day tutorial at the 2023 ACM SIGIR conference on Information Retrieval in Taipei: https://sigir.org/sigir2023/
- Inge Wortel moderated the course “AI in Perspectief” from the Radboud Academy on how AI impacts our life, which further featured contributions from Frederik Zuiderveen Borgesius, Martha Larson, and Tom Heskes.
- On March 23, the AI for Energy Grids Lab was officially launched. Within this lab, Charlotte Cambier van Nooten studies how graph neural networks can improve the efficiency and reliability of our energy grid under daily supervision of Yuliya Shapovalova and Tom van de Poll (Alliander).
- Tom Heskes contributed to the Studio Oost NL podcast on how small and medium-sized enterprises (SMEs) can get started with AI: https://www.ru.nl/ai/news-events/news/vm-eigen-news/studio-oost-nl-how-can-smes-get-started-ai/.
- Chris Kamphuis (with Faegheh Hasibi and Arjen de Vries) had his paper on entity annotations for the MS Marco collections accepted for 46th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR’23)! Furthermore, two full papers ("Cross-Market E-Commerce Question Answering by Jointly Ranking Questions and Products" by Negin Ghasemi, Arjen de Vries, Djoerd Hiemstra et al.; "Safe Deployment for Counterfactual Learning to Rank with Exposure-Based Risk Minimization" by Harrie Oosterhuis et al.) and a tutorial ("Recent Advances in the Foundations and Applications of Unbiased Learning to Rank" by Harrie Oosterhuis et al.) have been accepted at SIGIR'23 and will be presented in Taipei.
- The paper titled "Fairness Gerrymandering: Unfair, Undetectable Manipulation" by Tim de Jonge and Djoerd Hiemstra was accepted at the annual ACM Conference on Fairness Accountability and Transparency (ACM FAccT 2023).
- Harrie Oosterhuis presented a keynote talk titled: Counterfactual Learning to Rank for Search and Recommendation at the Fourth International Workshop on Algorithmic Bias in Search and Recommendation (BIAS 2023), which was held as part of the 45th European Conference on Information Retrieval (ECIR 2023): https://biasinrecsys.github.io/ecir2023/#keynote
- The data science group is covered in the April edition of the I/O magazine: https://ict-research.nl/wordpress/wp-content/uploads/2023/04/IO-magazine-NR1-2023.pdf Additionally, the edition features an article on Hannes Mühleisen and his DuckDB spin-off.