Dr.-Ing. Thomas Walther

RUB
Computer Science
INI
Alumni
Dr.-Ing. Thomas Walther

thomas.walther@rub.de

Computational Neuroscience

thomas.walther@rub.de

Research interests:

As a postdoctoral researcher in the 'Computational Neuroscience'
group, I focus on the modeling and understanding of cognition in mammals:
herein, I do specifically aim at unraveling cerebral circuits in the hippocampal
formation that participate in spatial navigation, learning, and memory extinction.
My further research interests span virtual reality systems, and robotics, computer
vision, computational auditory scene analysis, and Organic Computing.

Background:

I studied electrical engineering and information technology at the University of
Dortmund, finishing my studies there with my diploma thesis that focused on medical
image interpretation, tumor tracking and robot-assisted radiotherapy. My PhD work at the
'Institut für Neuroinformatik' (RUB) was on autonomous visual model learning, based on
principles of Organic Computing; this work resulted in my PhD in electrical engineering and
information technology granted by the Ruhr-Univesität Bochum.

2021

Emergence of complex dynamics of choice due to repeated exposures to extinction learning

Donoso, J. R., Packheiser, J., Pusch, R., Lederer, Z., Walther, T., Uengoer, M., et al.

Animal Cognition, 24(6), 1279–1297

@article{DonosoPackheiserPuschEtAl2021,
	author		=	{Donoso, José R. and Packheiser, Julian and Pusch, Roland and Lederer, Zhiyin and Walther, Thomas and Uengoer, Metin and Lachnit, Harald and Güntürkün, Onur and Cheng, Sen},
	title		=	{Emergence of complex dynamics of choice due to repeated exposures to extinction learning},
	journal		=	{Animal Cognition},
	volume		=	{24},
	number		=	{6},
	pages		=	{1279–1297},
	month		=	{May},
	year		=	{2021},
	doi		=	{10.1007/s10071-021-01521-4},
}

Context-dependent extinction learning emerging from raw sensory inputs: a reinforcement learning approach
Walther, T., Diekmann, N., Vijayabaskaran, S., Donoso, J. R., Manahan-Vaughan, D., Wiskott, L., & Cheng, S.
Scientific Reports, 11(1)

@article{WaltherDiekmannVijayabaskaranEtAl2021, author = {Walther, Thomas and Diekmann, Nicolas and Vijayabaskaran, Sandhiya and Donoso, José R. and Manahan-Vaughan, Denise and Wiskott, Laurenz and Cheng, Sen}, title = {Context-dependent extinction learning emerging from raw sensory inputs: a reinforcement learning approach}, journal = {Scientific Reports}, volume = {11}, number = {1}, month = {February}, year = {2021}, doi = {10.1038/s41598-021-81157-z}, }

Walther, T., Diekmann, N., Vijayabaskaran, S., Donoso, J. R., Manahan-Vaughan, D., Wiskott, L., & Cheng, S.. (2021). Context-dependent extinction learning emerging from raw sensory inputs: a reinforcement learning approach. Scientific Reports, 11(1). http://doi.org/10.1038/s41598-021-81157-z

2020

Automatic Tuning of RatSLAM′s Parameters by Irace and Iterative Closest Point
Menezes, M. C., Muñoz, M. E. S., Freitas, E. P., Cheng, S., Walther, T., Neto, A. A., et al.
In IECON 2020 The 46th Annual Conference of the IEEE Industrial Electronics Society (pp. 562–568)

@inproceedings{MenezesMuñozFreitasEtAl2020, author = {Menezes, M. C. and Muñoz, M. E. S. and Freitas, E. P. and Cheng, S. and Walther, T. and Neto, A. A. and Ribeiro, P. R. A. and Oliveira, A. C. M.}, title = {Automatic Tuning of RatSLAM′s Parameters by Irace and Iterative Closest Point}, booktitle = {IECON 2020 The 46th Annual Conference of the IEEE Industrial Electronics Society}, pages = {562–568}, month = {October}, year = {2020}, doi = {10.1109/IECON43393.2020.9254718}, }

Menezes, M. C., Muñoz, M. E. S., Freitas, E. P., Cheng, S., Walther, T., Neto, A. A., et al. (2020). Automatic Tuning of RatSLAM′s Parameters by Irace and Iterative Closest Point. In IECON 2020 The 46th Annual Conference of the IEEE Industrial Electronics Society (pp. 562–568). http://doi.org/10.1109/IECON43393.2020.9254718

2019

Deep reinforcement learning in a spatial navigation task: Multiple contexts and their representation
Diekmann, N., Walther, T., Vijayabaskaran, S., & Cheng, S.
In 2019 Conference on Cognitive Computational Neuroscience Berlin, Germany: Cognitive Computational Neuroscience

@inproceedings{DiekmannWaltherVijayabaskaranEtAl2019, author = {Diekmann, Nicolas and Walther, Thomas and Vijayabaskaran, Sandhiya and Cheng, Sen}, title = {Deep reinforcement learning in a spatial navigation task: Multiple contexts and their representation}, booktitle = {2019 Conference on Cognitive Computational Neuroscience}, publisher = {Cognitive Computational Neuroscience}, address = {Berlin, Germany}, year = {2019}, doi = {10.32470/CCN.2019.1151-0}, }

Diekmann, N., Walther, T., Vijayabaskaran, S., & Cheng, S.. (2019). Deep reinforcement learning in a spatial navigation task: Multiple contexts and their representation. In 2019 Conference on Cognitive Computational Neuroscience. Berlin, Germany: Cognitive Computational Neuroscience. http://doi.org/10.32470/CCN.2019.1151-0

2018

Unsupervised Acquisition of Human Body Models
Walther, T., & Würtz, R. P.
Cognitive Systems Research, 47, 68–84

Article Additional material

@article{WaltherWürtz2018, author = {Walther, Thomas and Würtz, Rolf P.}, title = {Unsupervised Acquisition of Human Body Models}, journal = {Cognitive Systems Research}, volume = {47}, pages = {68–84}, year = {2018}, doi = {10.1016/j.cogsys.2017.08.001}, }

Walther, T., & Würtz, R. P.. (2018). Unsupervised Acquisition of Human Body Models. Cognitive Systems Research, 47, 68–84. http://doi.org/10.1016/j.cogsys.2017.08.001

2017

Unsupervised Acquisition of Human Body Models using Principles of Organic Computing
Walther, T., & Würtz, R. P.
ArXiv e-prints

@article{WaltherWürtz2017, author = {Walther, Thomas and Würtz, Rolf P.}, title = {Unsupervised Acquisition of Human Body Models using Principles of Organic Computing}, journal = {ArXiv e-prints}, month = {April}, year = {2017}, }

Walther, T., & Würtz, R. P.. (2017). Unsupervised Acquisition of Human Body Models using Principles of Organic Computing. ArXiv e-prints. Retrieved from http://arxiv.org/abs/1704.03724

2020

Comparing Deep Reinforcement Learning Algorithms in Discrete and Continuous Environments
De Stefano, G.
Master’s thesis, Applied Computer Science, Ruhr University Bochum, Germany

@mastersthesis{De Stefano2020, author = {De Stefano, Gianluca}, title = {Comparing Deep Reinforcement Learning Algorithms in Discrete and Continuous Environments}, school = {Applied Computer Science, Ruhr University Bochum, Germany}, month = {December}, year = {2020}, }

De Stefano, G. (2020, December). Comparing Deep Reinforcement Learning Algorithms in Discrete and Continuous Environments. Master’s thesis, Applied Computer Science, Ruhr University Bochum, Germany.
Deep Reinforcement Learning in Virtual Reality Environments Driven by Physics Simulation
Woltersdorf, P.
Applied Computer Science, Ruhr University Bochum, Germany

@bachelorsthesis{Woltersdorf2020, author = {Woltersdorf, Philip}, title = {Deep Reinforcement Learning in Virtual Reality Environments Driven by Physics Simulation}, year = {2020}, }

Woltersdorf, P. (2020, October). Deep Reinforcement Learning in Virtual Reality Environments Driven by Physics Simulation. Applied Computer Science, Ruhr University Bochum, Germany.

2019

Reinforcement learning in multiple contexts: How does the network represent the learned information?
Diekmann, N.
Master’s thesis, Applied Informatics, Univ. of Bochum, Germany

@mastersthesis{Diekmann2019, author = {Diekmann, Nicolas}, title = {Reinforcement learning in multiple contexts: How does the network represent the learned information?}, school = {Applied Informatics, Univ. of Bochum, Germany}, month = {April}, year = {2019}, }

Diekmann, N. (2019, April). Reinforcement learning in multiple contexts: How does the network represent the learned information?. Master’s thesis, Applied Informatics, Univ. of Bochum, Germany.
Transfer of Learned Associations in a Reinforcement Learning Context
Vijayabaskaran, S.
Master’s thesis, Computational Engineering, Univ. of Bochum, Germany

@mastersthesis{Vijayabaskaran2019, author = {Vijayabaskaran, Sandhiya}, title = {Transfer of Learned Associations in a Reinforcement Learning Context}, school = {Computational Engineering, Univ. of Bochum, Germany}, month = {April}, year = {2019}, }

Vijayabaskaran, S. (2019, April). Transfer of Learned Associations in a Reinforcement Learning Context. Master’s thesis, Computational Engineering, Univ. of Bochum, Germany.

Emergent behavior and neural representations in spatial learning

Spatial navigation might appear to be a simple behavior, but closer inspection reveals that it is the complex result of many interacting sub-processes. We use deep reinforcement learning to understand how goal-directed behavior emerges in an artificial agent, how the deep neural network represents spatial information, and how the model's representations are related to neural codes for space in the hippocampus.

About the INI

The Institut für Neuroinformatik (INI) is a research unit of the Faculty of Computer Science at the Ruhr-Universität Bochum. Its scientific goal is to understand the fundamental principles through which organisms generate behavior and cognition while linked to their environments through sensory and effector systems. Inspired by our insights into such natural cognitive systems, we seek new solutions to problems of information processing in artificial cognitive systems. We draw from a variety of disciplines that include experimental psychology and neurophysiology as well as machine learning, neural artificial intelligence, computer vision, and robotics.

Contact

Universitätsstr. 150, Building NB, Room 3/32
D-44801 Bochum, Germany

Tel: (+49) 234 32-28967
Fax: (+49) 234 32-14210

sekretariat@ini.rub.de
www.ini.rub.de