CAIML Symposium 2025
The 4th CAIML Symposium will bring together leading experts and enthusiasts in AI to discover the future of the field.

May 19th 2025
- All day event.
- TU Wien, Campus Getreidemarkt Konferenzsaal TUtheSky
-
1060 Vienna, Getreidemarkt 9
Bauteil BA (Hoftrakt), 11. Stock, Raum BA11B09
Building on the momentum and successes of our previous gatherings, CAIML Symposium 2022, CAIML Symposium 2023 and CAIML Symposium 2024, we are excited to announce the upcoming 4th CAIML Symposium on May 19, 2025. This symposium will bring together leading experts and enthusiasts in the field of Artificial Intelligence and Machine Learning to discuss the latest advancements, challenges, and opportunities. Join us for a day filled with insightful talks, engaging discussions, and networking opportunities that promise to inspire and foster collaboration.
Program
- 9:00 – 09:15: Welcome address by Vice Rector Digitalisation and Infrastructure Wolfgang Kastner
- 9:15 – 10:15: Scientific talk: Sylvie Thiébaux, Australian National University and University of Toulouse
- 10:15 – 10:45: Coffee Break
- 10:45 – 12:15: iCAIML Doctoral College talks
- 12:15 – 14:00: Lunch Break & Networking
- 14:00 – 15:00: Scientific talk: Jilles Vreeken, Saarland University and CISPA Helmholtz Center for Information Security
- 15:00 – 15:15: Coffee break
- 15:15 – 16:45: Panel discussion “AI in Industry and Business: Current Applications and Future Challenges”
- 16:45 – 17:00: Coffee break
- 17:00 – 18:00: Scientific talk: Günter Klambauer, Johannes Kepler Universität Linz
Talks
Graph Learing for Planning
By Sylvie Thiébaux
State of the art methods for automated planning rely on heuristic state-space search. I will present recent work on graph representation learning to guide the search of automated planners. I will introduce graph neural network and other graph learning representations that exploit the relational structure of planning domains. They allow our planner GOOSE to learn heuristic cost estimates and state rankings from solutions to just a few small problems, and solve substantially larger problems than trained on. Perhaps surprisingly, our experimental results show that classical machine learning approaches vastly outperform deep learning ones in this context. Moreover, Greedy Best-First Search guided by our best learnt heuristics outperforms the state of the art model-based planner, Lama, on the problems of the latest International Planning Competition Learning track, leading to the possibility that learnt heuristics may replace existing model-based heuristics in the near future.
TBA
By Jilles Vreeken
A New Generation of Foundation Models Based on xLSTM
By Günter Klambauer
Foundation models, such as GPT and vision transformers, are almost all built on the famous Transformer architecture. Since about 2018 these transformers have replaced recurrent neural networks (RNNs) in natural language processing, computer vision, but also in other application areas. However, the computational costs of Transformers scale quadratically with sequence or context length which is one of the main reasons for the huge computational costs of AI systems across the world. xLSTM is a modern RNN, which is parallelizable similar to the Transformer , but its computational costs only scale linearly. Thus, xLSTM-based foundation models have the capacity to overtake Transformers as main components of AI systems. In this talk, we provide an overview of the development of AI systems from RNNs, to Transformers, LLMs and foundation models, and to the xLSTM architecture, its core component and its applications.
Speakers
Sylvie Thiébaux, Australian National University & University of Toulouse

Sylvie Thiébaux is a professor of computer science at The Australian National University, and a directrice de recherche and ANITI chair at the University of Toulouse. Her research interests are in artificial intelligence (AI), in particular in automated planning, scheduling, and search, their integration with optimisation, machine learning, and verification, as well as their applications to energy, manufacturing, and transport. She has received multiple academic and industry awards for her contributions to these areas. Sylvie is a fellow of the Association for the Advancement of Artificial Intelligence (AAAI) and editor in chief of the flagship journal Artificial Intelligence. She is a former councilor of AAAI, co-chair and president of the International Conference on Automated Planning and Scheduling (ICAPS), and director of the Canberra Laboratories of NICTA (now Data61).
Jilles Vreeken, Saarland University & CISPA Helmholtz Center for Information Security

Jilles Vreeken is tenured faculty at the CISPA Helmholtz Center for Information Security, where he leads the Exploratory Data Analysis group. In addition, he is Honorary Professor of Computer Science at Saarland University, Fellow of the ELLIS Society, and Faculty of the ELLIS Unit Saarbrücken.
His research interests include causal inference, machine learning, and data mining. He is particularly interested in developing well-founded theory and efficient methods for extracting informative causal models and patterns from large data, and putting these to good use. He has authored over 130 conference and journal papers. He received three best paper awards, the ACM SIGKDD 2010 Doctoral Dissertation Runner-Up Award, and the IEEE ICDM 2018 Tao Li Award.
He is program chair for IEEE ICDM 2025. Previously, he was a member of the ECML PKDD Steering Committee between 2016 and 2019 panel chair for SIAM SDM 2019, tutorial chair for SIAM SDM 2017, and program co-chair for ECML PKDD 2016. He co-organised ten workshops and co-lectured seven tutorials. He is associate editor of Knowledge and Information Systems (KAIS) and member of the editorial board of Data Mining and Knowledge Discovery (DAMI). In addition he regularly reviews for NeurIPS, ICML, KDD, AAAI, AISTATS, and so on.
He obtained his Ph.D. in Computer Science in 2009 from Universiteit Utrecht. Between 2009 and 2013 he was a post-doctoral researcher at Universiteit Antwerpen. Before joining CISPA in 2018, he was an independent research group leader (W2) at the DFG Cluster of Excellence on Multimodal Computing and Interaction (MMCI) and a Senior Researcher at the Max Planck Institute for Informatics.
Günter Klambauer, JKU Linz

Günter Klambauer is professor for “AI in Life Sciences” at the LIT AI Lab and the Institute for Machine Learning at the Johannes Kepler University Linz (JKU). He has been instrumental in the introduction of the “Artificial Intelligence” BSc&MSc program at the JKU, holds the central lectures “Deep Learning and Neural Networks” and heads the “Artificial Intelligence in Life Sciences” branch of study. He is founder of the ELLIS program “Machine Learning for Molecule Discovery” and was ELLIS Director of this program from 2022-2024. Since 2024, he is also research coordinator in Austria’s Cluster of Excellence in “Bilateral Artificial Intelligence” with a total funding of 33M€.
After studying mathematics and biology at the University of Vienna, Günter Klambauer began his research work in the field of machine learning and artificial intelligence at the Johannes Kepler University Linz where he obtained his doctorate in 2014. For his application of machine learning techniques in genetics and molecular biology, he was honored with the Austrian Life Science Award and the Award of Excellence of the Austrian Ministry of Science.
The “Self-normalizing neural networks” he developed were incorporated into Apple’s Siri speech recognition system in 2020, and major pharmaceutical companies use AI systems developed by his group. His current research focuses on developing broad AI systems for Life Sciences.