Welcome to the website of the Data and AI cluster at Eindhoven University of Technology!
We study foundations of data and AI for the present and the future. We design new methods, develop algorithms and tools with a view at expanding the reach of databases and AI and their generalization abilities. In particular, we study foundational issues of robustness, safety, fairness, trust, reliability, tractability, scalability, interpretability and explainability of data and AI. Currently, DAI includes five research groups: Uncertainty in AI, Generative AI, Automated ML, Data Mining, and Databases.
DAI research is regularly published at top data and AI journals and conferences such as IJCAI, AAAI, ICML, ECMLPKDD, ICDM, NeurIPS, VLDB, and SIGMOD among others. We actively collaborate with industry through multiple co-funded PhD projects.
Each academic year DAI attracts 70+ excellent MSc students for their internships and final thesis projects, many of which are done in collaboration with industry and other departments.
DAI is responsible for developing and teaching core data science and AI courses at MSc level, including Foundations of AI, Research Topics in Data Mining, Engineering Data Systems, Deep Learning, Reinforcement Learning, Text Mining, Uncertainty Representation and Reasoning, Generative Models, as well as challenge-based education such as Data Challenges, and at BSc level including Data Mining and Machine Learning, Datamodelling and databases and Responsible Data Science.
Shiwei Liu received the 2023 Best PhD Dissertation Runner-up Award from the Informatics Europe community and the Rising Star Award at the Conference on Parsimony and Learning (CPAL).In january of 2024, he will give a presentation at the University of Hong Kong (HKU).Lastly, he …
Our cluster will be greatly represented at NeurIPS! The following papers will be presented in New Orleans: Interpretable Reward Redistribution in Reinforcement Learning: A Causal Approach by Yudi Zhang, Yali Du, Biwei Huang, Ziyan Wang, Jun Wang, Meng Fang, Mykola Pechenizkiy Dynamic Sparsity Is Channel-Level Sparsity …
Hilde made it to the top 5 nominees for the Responsible AI Leader award from Women in AI Netherlands! Women in AI Netherlands describes a Responsible AI Leader as a woman leader, innovator, and visionary who is pioneering adopting responsible and ethical application of …
The Automated Machine Learning (AutoML) group explores how to use machine learning to learn how to create better machine learning models. The group focuses on bringing together AutoML, deep learning, meta-learning, transfer learning, continual learning, and other fields towards a single objective.More info View members
The data mining (DM) group is known for its contributions in predictive analytics, knowledge discovery and machine learning over evolving data streams. We study how to make machine learning models robust, trustworthy and explainable and how to (self-)audit AI solutions for their compliance to regulations and adherence to responsible AI guidelines, including aspects of privacy, fairness and non-discrimination among others.More info View members
The database (DB) group investigates data management and data-intensive systems, inspired by real-world application and analytics scenarios in close cooperation with public sector and industrial research partners. Expertise within the group includes query language design and foundations, query optimization and evaluation, data analytics, and data integration.More info View members
The Generative AI group focuses on building deep generative models (probabilistic modeling + deep learning) for defining generative processes, synthesizing new data, and quantifying uncertainty. The research carried out within the group is reinforced by applications in Life Sciences, Molecular Sciences, signal processing, smart devices, and smart apps (e.g., chatbots, art generation).More info View members
The UAI research group at TU/e explores uncertainty in AI and machine learning from multiple angles on principles of AI, theories of representation, probabilistic AI models, algorithms for learning, reasoning and decision making. There is also an important focus on approaches that are not only accurate but efficient, interpretative, robust and trustworthy.More info View members