MIN Faculty
Department of Informatics
Knowledge Technology

Dr. rer. nat. Dipl.-Inform. Stefan Heinrich

Postdoctoral Research and Teaching Associate of Knowledge Technology Group

Contact Info

Address: Universität Hamburg
Department of Informatics
Vogt Koelln Str. 30
22527 Hamburg, Germany
Office: F-208
Phone: +49 40 428 83 2516
Email: heinrich at informatik.uni-hamburg.de (PGP key)
Office Hours: by appointment

I am looking for highly-motivated students (Hiwi or BSc/MSc theses) to work with me on recurrent neural networks, machine learning, and natural language processing. If you are interested, please read this note and contact me via email.


Top

Research Interests and Background

My research interest is located in between artificial intelligence, cognitive psychology, and computational neuroscience. In my research I aim to explore the computational principles in the brain to foster our fundamental understanding of the brain's mechanisms but also to exploit them in developing intelligent systems. In particular, I look into the processes in the brain that form representations in multi-modal integration up to in language acquisition. As a central approach I model artificial neural networks, specifically continuous time recurrent neural networks (CTRNNs) and convolution neural networks (CNNs) on tasks in machine learning, natural language processing, and developmental robotics.

My current research project is entitled "Vision- and action-embodied language learning", a sub-project in the Sonderforschungsbereich/Transregio Project TRR 169 "Crossmodal Learning" funded by the DFG. The central question in this project is how we can build an artificial neural network model that processes auditory, visual and proprioceptive information and, thus, learns language by grounding speech in multi-modal embodied perception.

My previous education includes computer science and cognitive psychology on MSc level with a dregree from the University of Paderborn and computational modelling on PhD level with a degree from the University of Hamburg, but also programming and robot engineering als well as higher education teaching and training.


Top

Teaching

Currently I have no teaching obligations, but I will offer the following course via a teaching mandate:

WS 2016/17
[Show former teaching]

Nevertheless, I can supervise BSc and MSc theses in my fields of expertise. Please check the WTM BSc & MSc Thesis Offers for a general or pre-defined topic or contact me with your own ideas.

[Show completed theses]

I am the curriculum representative of Knowledge Technology Group.

For the BSc Informatik, I am part of the mentoring program.


Top

Short Curriculum Vitae

Since April 2016 Research Associate (TRR CML project) of Knowledge Technology Research Group, Department of Computer Science, University of Hamburg, Germany
2016 PhD degree in Computer Science (German Doktor der Naturwissenschaften/ Doctor rerum naturalium), Department of Computer Science, University of Hamburg, Germany
May 2010 - March 2016 Research and Teaching Associate of Knowledge Technology Research Group, Department of Computer Science, University of Hamburg, Germany
2009 Master degree in Computer Science (German Diplom-Informatiker), subsidiary subject: Psychology, Department of Computer Science, University of Paderborn, Germany
June 2006 - April 2010 Student Assistant as Developer and Programmer (PSI/REX Simulation Framework Project) of Institute of Polymer Engineering, University of Paderborn, Germany


Top

Theses

Natural Language Acquisition in Recurrent Neural Architectures
Dissertation, Supervisor: Prof. Dr. Stefan Wermter, Department of Computer Science, University of Hamburg, Germany, March/June 2016.
 
Memes and Cooperation in Multiagent Systems
Diploma Thesis, Supervisor: Prof. Dr. Hans Kleine Büning, Department of Computer Science, University of Paderborn, Germany, October 2009.  


Top

Publications

Alpay, T., Heinrich, S., Wermter, S. Learning Multiple Timescales in Recurrent Neural Networks. In Villa, A.E.P., et al., editors, Proceedings of the 25th International Conference on Artificial Neural Networks (ICANN 2016), pp. 132-139, Springer International Publishing. Hamburg, DE, September 2016.
 

Heinrich, S., Magg, S., Wermter, S. Analysing the Multiple Timescale Recurrent Neural Network for Embodied Language Understanding. In Koprinkova-Hristova, P.D., et al., editors, Artificial Neural Networks - Methods and Applications, SSBN 4, chap. 8, pp. 149-174, Springer International Publishing, 2015.
 


Heinrich, S., Wermter, S. Interactive Language Understanding with Multiple Timescale Recurrent Neural Networks. In Wermter, S., et al., editors, Proceedings of the 24th International Conference on Artificial Neural Networks (ICANN 2014), pp. 193-200, Springer International Publishing. Hamburg, DE, September 2014.
 


Twiefel, J., Baumann, T., Heinrich, S., Wermter, S. Improving Domain-independent Cloud-based Speech Recognition with Domain-dependent Phonetic Post-processing. In Brodley C.E. et al., editors, Proceedings of the 28th AAAI Conference on Artificial Intelligence (AAAI-14), pp. 1529-1535, AAAI Press. Québec, CA, 2014.
 


Rubruck, A., Aminian, A., Yalpi, J., Hanzal, P., Winde, S., Kuppusami, S., Younis, S., Thomas, S., Strahl, E., Bauer, J., Dávila-Chacón, J., Heinrich, S., Wermter, S. CoCoCo, Coffee Collecting Companion. In Brodley C.E. et al., editors, Proceedings of the 8th AAAI Video Competition 8th AAAI Conference on Artificial Intelligence (AAAI-14), AAAI Press. Québec, CA, 2014.
- Nominated for Best Robot Video Award
 


Heinrich, S., Weber, C., Wermter, S. Embodied Language Understanding with a Multiple Timescale Recurrent Neural Network. In Mladenov, V., et al., editors, Proceedings of the 23rd International Conference on Artificial Neural Networks (ICANN 2013), LNCS 8131, pp. 216-223, Springer Heidelberg. Sofia, BG, September 2013.
- Won ENNS Student Award
 


Heinrich, S., Folleher, P., Springstübe, P., Strahl, E., Twiefel, J., Weber, C., Wermter, S. Object Learning with Natural Language in a Distributed Intelligent System - A Case Study of Human-Robot Interaction. In Sun, F., editor, Proceedings of the IEEE First International Conference on Cognitive Systems and Information Processing (CSIP 2012), AISC 215, pp. 811-819, Springer Berlin. Beijing, CN, December 2012.
 


Heinrich, S., Weber, C., Wermter, S. Adaptive Learning of Linguistic Hierarchy in a Multiple Timescale Recurrent Neural Network. In Villa, A.E.P., et al., editors, Proceedings of the 22nd International Conference on Artificial Neural Networks (ICANN 2012), LNCS 7552, Part I, pp. 555-562, Springer Heidelberg. Lausanne, CH, September 2012.
 


Dávila-Chacón, J., Heinrich, S., Liu, J., Wermter, S. Biomimetic Binaural Sound Source Localisation with Ego-Noise Cancellation. In Villa, A.E.P., et al., editors, Proceedings of the 22nd International Conference on Artificial Neural Networks (ICANN 2012), LNCS 7552, Part I, pp. 239-246, Springer Heidelberg. Lausanne, CH, September 2012.
 


Heinrich, S., Wermter, S. Towards Robust Speech Recognition for Human-Robot Interaction. In Narioka, K., Nagai, Y., Asada, M., Ishiguro, H., editors, Proceedings of the IROS2011 Workshop on Cognitive Neuroscience Robotics (CNR), pp. 29-34, San Francisco, CA, USA, September 2011.
 


Heinrich, S., Eberling, M., Wermter, S. Determining Cooperation in Multiagent Systems with Cultural Traits. In Filipe, J., Fred, A., editors, Proceedings of the 3rd International Conference on Agents and Artificial Intelligence (ICAART 2011), Vol. 2, pp. 173-180, SciTePress. Rome, IT, January 2011.
 


Anwar, M. N., Oakes, M. P., Wermter, S. and Heinrich, S. Clustering Audiology Data. In Ramon, J., Vens, C., Driessens, K., Van Otterlo, M., Vanschoren, J., editors, Proceedings of the 19th Annual Belgian-Dutch Conference on Machine Learning (BeneLearn 2010), Leuven, BE, May 2010.



Top

Community Service

Program Committee of the International Conference on Artificial Neural Networks 2013, 2014, 2016.
 

Invited Talk: Natural Language Acquisition in Recurrent Neural Architectures. Tsinghua University, Department of Computer Science, Beijing, China, September 2016.
 

Local Organising Committee of the International Conference on Artificial Neural Networks 2014.
 

Local Media Interview: Rainer Hennig, Die Werkbank der Zukunft. LOG.Kompass, Issue 7/8-2013, DVV Media Group, July 2013.
 

Local Media Interview: Rainer Hennig, Bühne frei - für Roboter. LOG.Kompass, Issue 1/2-2013, DVV Media Group, January 2013.
 

Invited Talk: Effective and Adaptive Learning of CTRNNs for Multi-Modal Real-World Problems. CINACS Summer School 2012, Hamburg, Germany, September 2012.
 

Currently I support editing various conference proceedings (e.g. for ICANN) and journals (e.g. Journal of Advances in Artificial Intelligence, Journal of Behavioral Robotics, Biological Cybernetics, Transactions on Autonomous Mental Development). Further requests for reviews are welcome if the topics are matching my research interests!


Top

Further Affiliations and Collaborations

Earlier affiliations:
I sustain C3S