Retreat 2018 for the Max Planck ETH Center for Learning Systems
15.-17. October 2018, Insel Reichenau, Germany

We are pleased to announce this year’s retreat for the Max Planck ETH Center for Learning Systems. The retreat will take place at the Ganter Hotel Mohren, Priminstraße 141, Insel Reichenau, Germany. We cordially invite all CLS-affiliated researchers from ETH Zürich and MPI-IS to participate and actively contribute to this retreat and the development of CLS.

Participation: In order to optimally foster the CLS networking and exchange, participation is mandatory for the CLS doctoral fellows. Attendance is highly recommended to supervisors, full and associated members. Associated doctoral / postdoctoral fellows may attend if, for example, their area of research is otherwise underrepresented.

Aims and Scope

The retreat aims at fostering networking and exchange opportunities for CLS affiliated researchers while bringing together researchers from ETH Zürich and MPI-IS. The retreat will provide a platform to exchange ideas, present current research, discuss challenges and initiate future research collaborations within the Center among fellows, members and professors. This retreat will include all CLS topics such as learning control, robotics, haptic intelligence, machine learning, computer vision and perceiving systems.

The retreat is open to all CLS researchers from MPI-IS and ETH. We seek to create an informal atmosphere to foster open discussions and exchange of ideas.


Registrations can be submitted here

All fellows (full and associated) are asked to submit an abstract detailing their research background and goals. Requirements: the text must not exceed 600 words; the caption of your graph should have a maximum of 4 lines (about 400 characters). Do not cite more than 8 references. More information and guidelines on the abstract layout and content can be found here. Please use a horizontal (landscape) teaser image - E.g. 2:1 with a minimum image width of 800px and a resolution of 300 dots per inch (dpi).

Submissions will be reviewed based on scientific merit, writing and appearance. Some submissions will be selected for a plenary presentation of 20+10 minutes. All other accepted submissions will be presented in interactive poster format. In addition to presenting research results during their presentations, we encourage participants to also include open research questions and ideas that may lead to interesting and fruitful discussions.


In addition to these interactive presentations, keynote lectures and an exciting hackathon event, there will be plenty of time for discussions and informal interactions. We are excited to announce the following invited researchers:

Jonathan Fiene and Brian Wright, MPI-IS, Central Scientific Computing & Giambattista Parascandolo, MPI-IS, Empirical Inference: Machine Learning in Robotics Challenge.

Metin Sitti, MPI-IS, Physical Intelligence: Shape-programmable Magnetic Soft Milli/Microrobots.

Konrad Schindler, ETH Zürich, Department of Civil, Environmental and Geomatic Engineering: Inference, Learning and Attention Mechanisms that Exploit and Preserve Sparsity in CNNs.

Costs & logistics

Meals (lunch, dinner, coffee breaks) and accommodation for registered participants from ETH and MPI are covered by the Max Planck ETH Center for Learning Systems. There will be shuttle buses between the institutes ( ETH / MPI-TÜ / MPI-STG ) and the event venue.

All workshop activities will take place at the Hotel Ganter Mohren on the Island of Reichenau.

Ganter Hotel & Restaurant Mohren
Pirminstraße 141,
D-78479 Reichenau
Phone: +49 7534 9944 - 0
Fax: +49 7534 9944 - 610
E-Mail: hotel@mohren-bodensee.de

View map


Schedule (talks) & poster sessions

Schedule (overview)

October 15th (Monday)

Description From To
Arrival 10:45 11:15
Check-In 11:15 11:50
Lunch 12:00 13:30
(Abt Strabo Saal)
13:30 14:00
Keynote in Robotics: Metin Sitti: Shape-programmable Magnetic Soft Milli/Microrobots
(Abt Strabo Saal)
14:00 15:00
Machine Learning in Robotics Challenge (with Coffee Break at 16:30)
by Giambattista Parascandolo, Jonathan Fiene & Brian Wright
(Abt Strabo Saal)
15:00 18:30
Break 18:30 19:00
Dinner 19:00 20:30
Round-Table Scientific Discussions
(Abt Strabo Saal)
20:30 22:00

October 16th (Tuesday)

Description From To
Breakfast 07:00 09:00
Launching new Initiatives and Collaborations (with Coffee Break at 10:45)
(Hotel Entrance)
09:00 12:30
Lunch 12:30 14:00
Parallel Session Seminars (with Coffee Break at 15:30) 14:00 17:00
Seminar: Machine Learning (Abt Strabo Saal)
Seminar: Physical Intelligence, Haptic Intelligence & Robotics (Konferenzraum 1)
Seminar: Perceiving Systems & Computer Vision (Konferenzraum 2)
Break 17:00 18:00
Keynote in Computer Vision: Konrad Schindler: Inference, Learning and Attention Mechanisms that Exploit and Preserve Sparsity in Convolutional Neural Networks
(Abt Strabo Saal)
18:00 19:00
Dinner 19:00 20:30
Poster Session
(Abt Strabo Saal)
20:30 22:00

October 17th (Wednesday)

Description From To
Breakfast 07:00 09:00
Parallel Session Seminars (with Coffee Break at 10:30) 09:00 12:00
Seminar: Fellows: Planning the Future of CLS (Abt Strabo Saal)
Seminar: Members: Planning the Future of CLS (Konferenzraum 1)
Lunch 12:30 13:45
Departure 13:45 14:00



Barbara Kreilkamp

Coordinator for the Max Planck ETH Center for Learning Systems

Hakan Ceylan

MPI-IS, Physical Intelligence Department

Sandra Lang

ETH Zürich, Institute for Machine Learning