Content:

How do humans and animals learn from experience, make decisions under uncertainty, form internal models of the world, and adapt their behavior to changing environments? How can similar problems be approached in artificial systems and computational models?

This seminar examines learning and decision making from the perspectives of computational neuroscience, cognitive science, reinforcement learning, and artificial intelligence. A central goal of the seminar is to compare how biological and artificial systems solve related computational problems, including reward learning, planning, exploration, memory, prediction, and adaptive behavior.

Topics include reinforcement learning, dopamine and reward prediction errors, model-free and model-based decision making, cognitive maps, memory replay, curiosity and exploration, planning under uncertainty, attention and cognitive control, and computational models of learning and behavior. The seminar will combine classical and modern research papers as well as review articles from neuroscience, cognitive science, and artificial intelligence.

The seminar is organized around thematic topic groups that contrast biological and artificial approaches to similar problems, such as on dopamine signaling and temporal-difference learning, hippocampal replay and experience replay, or cognitive maps and model-based reinforcement learning.

Each student will present and discuss one or more research papers related to a seminar topic. Presentations will be followed by an active discussion involving the entire seminar group.

Max. number of participants: 20

Learning Outcomes:

• Knowledge of computational approaches to learning and decision making in biological and artificial systems
• Ability to explain and critically discuss computational models and experimental findings from scientific research papers
• Insight into how brains and artificial systems solve problems such as prediction, planning, exploration, and adaptive behavior
• Ability to compare biological and artificial approaches to similar computational problems
• Practice in presenting and discussing scientific literature in an interdisciplinary setting

Examination:

Seminar contribution

Requirements for the awarding of credit points:

Seminar presentation, active participation

Enrollment:

The number of participants is limited. From the 1st to the 31st of August 2026 interested students can apply for a place in the seminar. The allocation of places will take place via the following central Moodle course of the faculty: https://moodle.ruhr-uni-bochum.de/course/view.php?id=62179. 

The final allocation of places will take place by the 16th of September 2026 at the latest. Complete your binding registration by registering for the seminar via FlexNow. Information on the deadlines can be found on the website of the Office of Academic Affairs of the Faculty of Computer Science. Please note that following the above steps is mandatory. Enrollment via FlexNow without prior registration via the central Moodle course is not permitted.

Learning Outcomes:
• Knowledge on different algorithms and computational approaches for decision making
• Explain the underlying mathematical problem formulations and the implementation of the algorithms to solve them
• Insight into different types of uncertainty and the balancing of multiple objectives
• Discuss practical applications of the theoretical frameworks
• Present the algorithms and mathematical problem formulations to an audience

Examination:
Oral presentation

Prof. Dr. Robert Schmidt Lecturer

(+49) 234-32-27300 robert.schmidt@rub.de NB 3/68

Course type

Seminars

Credits

3

Term

Winter Term 2026/2027

E-Learning

moodle course available

Seminar

Takes place every week on Monday from 16:00 to 18:00.
First appointment is on 12.10.2026
Last appointment is on 01.02.2027

Knowledge of calculus, linear algebra, and probability concepts. Background in artificial intelligence, e.g. via the course “Introduction to Artificial Intelligence”.