Modeling spatial behaviors and representations using deep reinforcement learning Computational Neuroscience


Spatial navigation is the ability of organisms to extract and move towards a specific goal or location within their environment. This is a complex cognitive process that involves many different behaviors and strategies (Parra Barrero et al., 2023). We use reinforcement learning (RL) to model different aspects of navigation by training agents to learn to solve different navigation tasks through trial and error. Since RL operates in a closed loop, we can study the interaction of behavior and representations in artificial agents as they learn to solve tasks. To this end, we develop and maintain the software package CoBeL-RL (Diekmann et al., 2023), which provides tools to easily simulate a variety of scenarios and RL agents.

Within this framework, we offer bachelor's and master's theses. Depending on the level (bachelor or master) and preferences of the student, the project can be more technically oriented (developing new features for CoBeL-RL within the scope of answering a scientific question) or more an in-depth exploration of how artificial agents navigate and comparing them to navigation in animals.


Python programming skills are required.


Parra-Barrero, E., Vijayabaskaran, S., Seabrook, E.,  Wiskott, L., & Cheng, S. (2023). A map of spatial navigation for  neuroscience. In Neuroscience & Biobehavioral Reviews (Vol. 152,  p. 105200). Elsevier BV.

Diekmann, N., Vijayabaskaran, S., Zeng, X., Kappel, D., Menezes, M. C.,  & Cheng, S. (2023). CoBeL-RL: A neuroscience-oriented simulation  framework for complex behavior and learning. In Frontiers in  Neuroinformatics(Vol. 17). Frontiers Media SA.


Prof. Dr. Sen Cheng and Sandhiya Vijayabaskaran (

The Institut für Neuroinformatik (INI) is a central research unit of the Ruhr-Universität Bochum. We aim to understand the fundamental principles through which organisms generate behavior and cognition while linked to their environments through sensory systems and while acting in those environments through 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 approaches from psychology and neurophysiology as well as theoretical approaches from physics, mathematics, electrical engineering and applied computer science, in particular machine learning, artificial intelligence, and computer vision.

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

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