Seminar: Intrinsically Motivated & Object-Centric RL

We discuss recent papers on curiosity, world models, object-centric perception for control, hierarchical and model-based RL. Each student presents one recent paper (25–30 min) and submits a short write-up.

• Who: ML Master, advanced BSc with ML background
• How to apply: please send us an email with the subject “Seminar RL” and 2–3 preferred papers – see the contact page for the address and details.

Seminar: ML for Science, Robotics & Embodiment

Focus on equation learning, symbolic regression, ML for physics/biology/neuroscience data, learned control for embodied agents, and haptic sensing.

• Format: kick-off, intermediate feedback, final mini-workshop
• ECTS: 4–6 (depending on faculty rules)

Participation in a seminar is a plus if you later ask for a thesis with us.

For the email address and what to include, please refer to the How to apply / contact us page.