Research Interests at the Institute

Algebraic and Logical Foundations of Computer Science

Our Research at the Chair of Algebraic and Logical Foundations of Computer Science addresses formal methods for the design and analysis of parallel systems, and related foundations on automata theory and logic.

Main research topics are algorithms for the quantitative analysis of probabilistic models (“probabilistic model checking”) against temporal specifications and various forms of cost-utility requirements, abstraction and reduction techniques, symbolic methods, temporal logic, composition operators and semantical aspects of modelling and coordination languages. The development of the theoretical foundations is accompanied by experimental studies with prototypical implementations. The feasibility of probabilistic model checking tools for the analysis of energy-aware computing techniques and heterogeneous systems is studied among others within the collaborative research center Highly Adaptive Energy-Aware Computing  (HAEC) and the center for Advancing Electronics Dresden (cFAED).

Our teaching conveys the foundations of theoretical computer science for undergraduates and various courses on model checking and logic for advanced studies.

Automata Theory

The research at the chair of automata theory is mostly concered with answering foundational questions about logic-based knowledge representation formalisms.  A particular focus lies on the examination of newly designed description logics, their expressive power and their reasoning complexity.  Other topics related to description logics involve the investigation of unification in description logics as well as the semi-automatic generation of description logic knowledge bases.  For the latter, connections to the theory of formal concept analysis are exploited.

Another research topic represented at the chair of automata theory is molecular computing.

Foundations of Programming

The Chair for Foundations of Programming focuses its research on the
theory of weighted tree automata and its application to natural language
processing (NLP), in particular, statistical machine translation. We
investigate algorithms for the extraction of automata from large corpora
and for the training of probabilities to their transitions. To showcase
the manifold applications of the theoretical results in NLP, we have
developed the system Vanda, an integrated development environment for
the design of NLP algorithms. With Vanda, we pursue a modular approach,
resting on three columns: a mathematical framework based on initial
algebra semantics that allows expressing a wide variety of grammars and
automata, an implementation of this framework in Haskell, and a
workflow-based user interface to the implementation.

Knowledge-Based Systems

The research group Knowledge-Based Systems is concerned with methods for the intelligent management and processing of information in computer systems. This includes research questions from knowledge representation, reasoning and formal logic, but also covers topics related to databases and distributed systems. Important application areas of this research can be found in the fields of semantic technologies, artificial intelligence, and knowledge management.