Find Academic Supervisors

Robert Peharz is an Assistant Professor at Graz University of Technology, where he leads research at the Institute of Machine Learning and Neural Computation. His work focuses on probabilistic machine learning, with particular emphasis on tractable probabilistic models, causality, and neurosymbolic AI. Education and Career PhD from TU Graz (Austria) in 2015 Postdoc at Medical University of Graz Postdoc and Marie-Curie Individual Fellow at University of Cambridge (2017-2019) Assistant Professor at Eindhoven University of Technology (2019-2021) Current: Assistant Professor at Graz University of Technology Research Interests Peharz's research spans multiple areas of artificial intelligence with a focus on making probabilistic reasoning both theoretically sound and practically efficient. His work addresses fundamental challenges in tractable probabilistic inference and learning, probabilistic circuits as a unified framework for deep generative models, Bayesian causal inference, and neurosymbolic AI combining sub-symbolic and symbolic approaches. His research has applications in cybersecurity, healthcare, and energy systems. Research Projects VENTUS (2024-present): Physics-informed, probabilistic and causal machine learning for wind energy systems NEO DNA (2023-present): DNA-based data storage systems using computer vision and probabilistic ML VanillaFlow (2023-present): AI-guided development of novel vanillin-based molecules for redox flow batteries Bilateral AI : Cluster of Excellence focused on Broad AI combining sub-symbolic and symbolic AI approaches Awards and Recognition Finalist for TUG's Excellent Teaching Award (2023) for all 3 of his courses Marie-Curie Individual Fellow at University of Cambridge Academic Service Peharz is actively involved in the academic community through conference organization and reviewing: Area Chair: UAI (2022), ECML/PKDD (2022) Senior Committee Member: UAI (2021), IJCAI (2019, 2020) Reviewer for major conferences including ICML, NeurIPS, AAAI, IJCAI-ECAI Teaching and Mentorship Peharz supervises multiple PhD students working on diverse projects at the intersection of machine learning, causality, and neurosymbolic AI. His current advisees include Sepideh Adamiat, Irina Dobrianski, Johannes Exenberger, Giacomo Di Gobbi, Tim d'Hondt, Christian Toth, and Thomas Wedenig. Previous students include Alvaro Correia, Martin Trapp, and David Montalvan.

• Probabilistic Machine Learning
• Tractable Probabilistic Models
• Artificial Intelligence
• +6 more

Matthias Paul Lanzinger is an Assistant Professor at the Technische Universität Wien's Faculty of Informatics, Department of Database and Artificial Intelligence. His research focuses on algorithms, graph neural networks, hypergraph decomposition techniques, parameterized complexity, and computational logic. He leads projects like 'DeConquer' (Vienna Science Fund) and 'HyperTrac', exploring efficient query processing and hypergraph-based algorithms. Research interests include theoretical computer science, database systems, and applying logical frameworks to solve complex computational problems. Recent work emphasizes hypertree decompositions, fuzzy Datalog, and graph motif analysis via the Weisfeiler-Leman test. He co-edited the 2024 Datalog-2.0 workshop proceedings and has supervised students on topics like column-store performance and graph query languages. His publications span venues like ACM Transactions on Database Systems, ICLR, and IJCAI, highlighting contributions to algorithmic efficiency, database theory, and logical reasoning systems. Active in academic service, he teaches courses on database systems, scientific research, and advanced topics in informatics.

• Algorithms
• Graph Neural Networks
• Problem decomposition techniques
• +4 more

Jan Niclas Dreier is a PostDoc Researcher at TU Wien's Department of Algorithms and Complexity. His research focuses on theoretical computer science, particularly algorithmic meta-theorems, structural graph theory, and computational complexity. He is affiliated with projects such as REVEAL-AI and SLIM. His work addresses topics like monadic stability, SAT solving, and model checking on sparse graph classes. Research Interests : Dreier's primary areas include algorithm design, parameterized complexity, and logic in computer science. His contributions span model checking for graph classes, backdoor analysis in SAT solving, and combinatorial properties of monadic dependence. Recent work explores the interplay between structural graph theory and algorithmic efficiency, with applications to finite model theory and pseudorandom models. Teaching : He teaches courses such as Algorithmic Meta-Theorems , Algorithmics , and Bachelor Thesis in Computer Science . His courses emphasize theoretical foundations and practical algorithmic techniques. Projects : Active involvement in the REVEAL-AI project (2020–2024) and SLIM (2019–2024), focusing on algorithmic advancements in AI and structural graph analysis.

• Algorithmics
• Computational Complexity
• Graph Theory
• +3 more