The position is forseen to do research and development in the BMBF-funded project OKS. This project aims at the optimization of human-machine interaction with learning conversational speech interfaces. Machine learning is already successfully used in research for the implementation of intelligent and natural speech-based personal assistants. However, even large ICT enterprises do not over automatic methods for the generation of dialog systems.
The aim of the project is to fill this implementation gap with tried and tested tools and work processes for the German economy in order to enable the cost-efficient development of such speech-based assistants. To this end, adaptive dialogue systems shall be created automatically from knowledge bases and optimized by using machine learning, especially reinforcement learning. On the basis of this technology, systems will be created that enable speech access to complex data structures in everyday professional life and speech-based support in everyday private life. The solutions available in basic research have to be validated with regard to their usage for dialog system architectures and evaluated with users.
If you have the wish and the right suitability, you can pursue a doctorate at our chair.
- The concrete tasks include, depending on previous knowledge:
Derivation of dialog domains and dialog managers from ontologies and database systems
- Conception of dialogue strategies and dialogue systems
- Implementation of a reinforcement learning environment for statistical dialogue managers (e.g. based on Q-Learning or neural networks)
- Conception of suitable Reward functions
- Implementation of heuristic and data-driven user simulations
- Integration of the dialog manager in spoken dialog systems
- Evaluation of spoken dialog systems
- The opportunity for working on a doctoral dissertation is given.
- Successfully completed university degree (Master, Diplom or equivalent) in computer science, computational linguistics or a comparable field of study
- Good programming skills in Python and secure handling of GIT as well as unit tests, ideally experience with Continuous Integration
- Good knowledge in dealing with ontologies, SQL or No-SQL databases
- Interest in conducting empirical studies with test participants or expert evaluations for the evaluation of spoken dialogue systems
- Language skills: German at the level of a native speaker, English fluent in spoken and written form
- Good skills in the field of machine learning (training and evaluation), ideally experience with reinforcement learning
- Ideally prior experience in the development of spoken dialog systems or natural language processing
- A desire to work in an international and interdisciplinary environment
How to apply:
Please send your written application with the reference number and the usual documents (in particular letter of application, curriculum vitae, copies of certificates, job references) to Technische Universität Berlin – Der Präsident – Fakultät IV, Institut für Softwaretechnik und Theoretische Informatik, Quality and Usability Lab, Prof. Dr. Möller, Sekr. TEL 18, Ernst-Reuter-Platz 7, 10587 Berlin or by e-mail to email@example.com.
To ensure equal opportunities between women and men, applications by women with the required qualifications are explicitly desired.
Qualified individuals with disabilities will be favored. The TU Berlin values the diversity of its members and is committed to the goals of equal opportunities.
Please send copies only. Original documents will not be returned.
The offer is under the reserve that funds are granted – part-time employment may be possible.
About Technische Universität Berlin, Quality and Usability Lab
The majority of systems and services that are provided by computer science, electrical engineering and information technology finally are oriented towards the needs of their human users. To successfully build such systems and services it is essential to investigate and understand users and their behavior when interacting with technology. On the hand, this understanding allows to derive design principles for human-machine-interfaces and to define requirements concerning the system and its underlying technologies. On the other hand, new available technologies do offer new possibilities regarding the design of such interfaces, and the development of new kinds of interaction.
The Quality and Usability Lab is part of TU Berlin’s Faculty IV for Electrical Engineering and Computer Science and deals with the design and evaluation of human-machine interaction. Here, the main subjects of our research are human perception, aspects of technical system that are related to the interaction, as well as interaction design.