Hauptinhalt

Topinformationen

Personen

Reading group on natural and artificial reinforcement learning

8.30702

Dozenten

Beschreibung

Content and goal:
In this course, you will explore research on theories, neural evidence, and successful implementations of reinforcement learning - the workhorse underlying how all agents - biological or artificial - learn. On a weekly basis, we will discuss classic and newly emerging research articles decided by the instructors and decided amongst the group during the course. Each student, or group of students, depending on the number of students enrolled, will be required to prepare and present at least one paper, to guide the group through it and to lead the in-depth discussion. By the end of the course, you will have a wide understanding of the common conceptualizations and trends in this direction of research. You will be better at parsing papers - getting to the central messages quickly and assessing the validity of the claims being made.

Target audience:
Master students who have at least completed an introductory course in cognitive neuroscience or neurobiology and an introductory course to deep learning. Highly motivated bachelors students are welcome. Bear in mind that the course content will be complex.

Grading:
The course will be completed with a written report, to be handed in by each student, in which you will be asked to answer a guiding question or review a common theme among the papers covered. The course grade will be partly based on the weekly presentations and the report.

Weitere Angaben

Ort: 69/E23
Zeiten: Do. 16:00 - 18:00 (wöchentlich)
Erster Termin: Donnerstag, 17.04.2025 16:00 - 18:00, Ort: 69/E23
Veranstaltungsart: Seminar (Offizielle Lehrveranstaltungen)