Founded in 2017 by Holger Hoos, Professor of Machine Learning at the Leiden Institute of Advanced Computer Science, the ADA Research Group pursues the development of AI (artificial intelligence) techniques that complement, rather than replace, human intelligence. In particular, our research is focussed on methods for the automated design and analysis of algorithms for computationally challenging problems, leveraging human creativity, advanced machine learning and optimisation methods, and lots of compute cycles. We work on a broad range of problems, including propositional satisfiability (SAT), AI planning, mixed integer programming (MIP), the travelling salesperson problem (TSP), supervised and semi-supervised machine learning, as well as a range of real-world applications.
Our group is particularly well known for its work on programming by optimisation (PbO) and automated machine learning (AutoML). Much of what we do can be seen as generalised machine learning, aimed at automating the construction of algorithms for arbitrary computational problems.
What does 'ADA' stand for?
- Automated Design of Algorithms (which includes, as a special case, machine learning and AutoML)
- Analysis and Design of Algorithms (mostly using empirical methods, based on machine learning and optimisation)
- ADA disrupts AI (a nested acronym which can be expanded in many ways)
- Ada Lovelace, whose vision continues to inspires us
ADA @ LIACS and ADA @ UBC
Currently, the ADA Research Group has two branches: One at the Leiden Institute of Advanced Computer Science (LIACS) in Leiden, The Netherlands, and the other at the University of British Columbia (UBC) in Vancouver, Canada. We also have active research collaborations with various other institutions, including Université Libre de Bruxelles, Belgium; Albert-Ludwigs-Universität Freiburg, Germany; Westfälische Wilhelmsuniversität Münster, Germany; Università degli Studi di Brescia, Italy; and Université de Lille, France.