Anna Rapberger: “Learning Assumption-Based Argumentation Frameworks from Examples”
Join us for a talk with Anna Rapberger.
May 7th 2026
- 12:15 – 13:00 CEST
- TU Wien, FAV EG C (Seminarraum Gödel)
-
Favoritenstraße 9-11, 1040 Vienna
Raum HBEG10
On This Page
About the Event
May 07, 2026
12:15 – 13:00 PM
(12:15) CEST
A map to the location can be found at TU Maps.
Barrier free access is available through the entrance at Favoritenstraße 9.
Abstract
What can argumentative outcomes tell us about the rules that generated them? This talk studies the problem of learning defeasible inference rules from examples through the lens of computational argumentation. We base our investigations on Assumption-Based Argumentation (ABA), a versatile structured argumentation formalism which has shown to capture many non-monotonic reasoning formalisms, including answer set programming and Dung’s abstract argumentation frameworks. In our learning setting, we assume background knowledge in the form of an ABA framework together with new information in the form of partial extensions or labellings. Our focus is on the theoretical properties and limits of this setting: we identify conditions under which solutions exist, investigate the verification and optimality of such solutions, and analyse the computational complexity of the resulting decision problems.
About the Speaker
Anna Rapberger is a Postdoc at TU Dortmund and a Honorary Research Fellow at Imperial College London. At TU Dortmund, she is part of the group Knowledge Based Systems, led by Jean Jung.
She completed her PhD in March 2023 as part of the Doctoral College Logical Methods in Computer Science (LogiCS) at TU Wien, supervised by Stefan Woltran. Between 2023-2025, she was a research associate at Imperial College London and a member of the Computational Logic and Argumentation group (CLArg), led by Francesca Toni.
Her research centers around, broadly speaking, symbolic AI, particularly in the area of knowledge representation and non-monotonic reasoning. Her research on theoretical and practical aspects of computational argumentation was recognized with the KR Early Career Award 2024 (press). Her research interests furthermore include causal discovery, explainable AI, bias detection through explainable AI, logic programming, and non-classical logics.