Search

conference
01/09/2014 04/09/2014

Autonomous Learning

Autonomous Learning research aims at understanding how autonomous systems can efficiently learn from the interaction with the environment, especially by having an integrated approach to decision making and learning, allowing systems to autonomously decide on actions, representations, hyperparameters and model structures for the purpose of efficient learning.

In this summer school international and national experts will introduce to the core concepts and related theory for autonomous learning in real-world environments. We hope to foster the enthusiasm of young researchers for this exciting research area, giving them the opportunity to meet leading experts in the field and similarly interested students. Our school offers an opportunity to look into fundamental and advanced aspects of autonomous learning. The tutorials are structured around three themes:

  1. learning representations,
  2. acting to learn (exploration), and
  3. learning to act in real-world environments.

The themes include but are not restricted to the following subjects:

Theme 1: Learning representations

  • compressed sensing/sparse coding
  • deep learning
  • hierarchical representations in perception
  • learning abstractions and symbols

Theme 2: Acting to learn (exploration)

  • foundations of optimal exploration and information seeking
  • approaches to intrinsic motivation

Theme 3: Learning to act in real-world environments

  • relational reinforcement learning
  • Bayesian cognitive robotics

Financial support

The summer school is financially supported by the autonomous-learning.org - external>DFG Priority Program Autonomous Learning.

Application

Registration is possible with the application form on this website until May 31. We will ask you for a short CV with a description of your research interest and a reference contact (we will email your recommender to provide a brief letter). Applicants will be informed of the outcome of their application until June 17.

Accommodation and costs

This course is free of charge, but participants have to cover their own travel, room and board. We have blocked an affordable room allocation at A&O Hostel in Leipzig (Hauptbahnhof, Brandenburger Straße 2, 04103 Leipzig). More information by email after your application was accepted.

Speakers

Shun-ichi Amari

Tamim Asfour

Michael Beetz

Matthias Bethge

Christos Dimitrakakis

Keyan Ghazi-Zahedi

Thomas Martinetz

Helge Ritter

Satinder Singh

Friedrich Sommer

Marc Toussaint

Program

10:30 - 11:00
11:00 - 12:30
Theme 1: Learning representations :
Shun-ichi Amari (RIKEN, Japan)
Information geometry and its applications to learning

12:30 - 14:30
14:30 - 16:00
Theme 1: Learning representations:
Shun-ichi Amari (RIKEN, Japan)
Basic principles of supervised and unsupervised learning: toward understanding deep learning

17:30 - 00:00
09:00 - 10:30
Theme 3: Learning to act in real-world environments:
Christos Dimitrakakis (Chalmers University of Technology, Gothenburg, Sweden)
Planning under uncertainty: Markov decision processes

10:30 - 11:00
11:00 - 12:30
Theme 2: Acting to learn (exploration) :
Keyan Ghazi-Zahedi (Max Planck Institute for Mathematics in the Sciences, Germany)
Intrinsically motivated exploration of behavioural modes

12:30 - 14:30
14:30 - 16:00
Theme 1: Learning representations:
Thomas Martinetz (University of Lübeck, Germany)
Sparse coding and efficient sensing

16:00 - 16:30
16:30 - 18:00
Theme 2: Acting to learn (exploration) :
Friedrich Sommer (Redwood Center for Theoretical Neuroscience, UC Berkeley, USA)
Information-theory based policies for learning in (embodied) closed sensori-motor loops

09:00 - 10:30
Theme 3: Learning to act in real-world environments :
Tamim Asfour (KIT Karlsruhe, Germany)
Structural bootstrapping for 24/7 humanoids

10:30 - 11:00
11:00 - 12:30
Theme 2: Acting to learn (exploration) :
Marc Toussaint (Stuttgart University, Germany)
Bandits, global optimization, active learning, and Bayesian reinforcement learning -- understanding the common ground

12:30 - 14:30
14:30 - 17:00
18:00 - 00:00
09:00 - 10:30
Theme 3: Learning to act in real-world environments:
Michael Beetz (Bremen University, Germany)
Autonomous Learning for Human-scale Everyday Manipulation Tasks

10:30 - 11:00
11:00 - 12:30
Theme 3: Learning to act in real-world environments :
Helge Ritter (Bielefeld University, Bielefeld, Germany)
Issues, algorithms, and challenges

12:30 - 14:30
14:30 - 16:00
Theme 1: Learning representations:
Matthias Bethge (University of Tübingen, MPI for Biological Cybernetics, Bernstein Center for Computational Neuroscience, Germany)
Natural image statistics & neural representation learning

16:00 - 16:30
16:30 - 18:00

Participants

Laith Alkurdi

Technische Universität München (Muenchen), Germany

Shun-ichi Amari

RIKEN, Japan

Tamim Asfour

KIT Karlsruhe, Germany

Nihat Ay

Max Planck Institute for Mathematics in the Sciences (Leipzig), Germany

Michael Beetz

Bremen University, Germany

Fabien Benureau

Inria Bordeaux (Talence), France

Oswald Berthold

Humboldt-Universität zu Berlin, Germany

Matthias Bethge

University of Tübingen, MPI for Biological Cybernetics, Bernstein Center for Computational Neuroscience, Germany

Dieter Büchler

Max Planck Institute for Intelligent Systems/TU Darmstadt, Germany

Christos Dimitrakakis

Chalmers University of Technology (Gothenburg), Sweden

Damien Drix

BCCN / Humboldt Universität zu Berlin, Germany

Peter Englert

Universität Stuttgart, Germany

Yanwei Fu

Queen Mary University of London, United Kingdom

Tim Genewein

Max Planck Institute for Biological Cybernetics (Tübingen), Germany

Keyan Ghazi-Zahedi

Max Planck Institute for Mathematics in the Sciences, Germany

Simon Hangl

University of Innsbruck (Pfunds), Austria

Raphael Holca-Lamarre

Technical University of Berlin, Germany

Federica Inderst

Università degli Studi di Roma 3 (Rome), Italy

Marika Kaden

UAS Mittweida, Germany

Martin Knopp

Technische Universität München, Germany

Okan Koç

MPI Tuebingen, Germany

Robin Lamarche-Perrin

Max Planck Institute for Mathematics in the Sciences (Leipzig), Germany

Marion Lange

Stuttgart University / TU Berlin (Stuttgart / Berlin), Germany

Robert Lieck

Universität Stuttgart, Germany

Qi Liu

University of Kaiserslautern, Germany

Tamas Madl

University of Manchester, United Kingdom

Guido Manfredi

LAAS-CNRS (Toulouse), France

Dimitrije Markovic

Max Planck Institute for Human Cognitive and Brain Sciences (Leipzig), Germany

Thomas Martinetz

University of Lübeck, Germany

Carlos Mastalli

Istituto Italiano di Tecnologia (Genoa), Italy

Dominik Meyer

Technische Universität München, Germany

Clément Moulin-Frier

INRIA Bordeaux Sud-Ouest, France

Hung Ngo

Swiss AI Lab IDSIA, University of Lugano (Manno-Lugano), Switzerland

Benjamin Paassen

CITEC - Bielefeld University, Germany

Paolo Perrone

Max Planck Institute for Mathematics in the Sciences (Leipzig), Germany

Anja Philippsen

Bielefeld University, Germany

Alexander Priamikov

Frankfurt Institute for Advanced Studies (Frankfurt am Main), Germany

Alexander Rietzler

University of Innsbruck, Austria

Helge Ritter

Bielefeld University, Germany

Hans-Christian Ruiz

Radboud University (Nijmegen), Netherlands

Christopher Schindlbeck

Leibniz Universität Hannover, Germany

Henry Schütze

Universität zu Lübeck, Germany

Satinder Singh

University of Michigan, USA

Sacha Sokoloski

Max Planck Institute for Mathematics in the Sciences (Leipzig), Germany

Friedrich Sommer

Redwood Center for Theoretical Neuroscience, UC Berkeley, USA

Sebastian Stabinger

University Innsbruck, Austria

Marc Toussaint

Stuttgart University, Germany

Christian Wirth

TU Darmstadt, Germany

Noga Zaslavsky

The Hebrew University of Jerusalem, Israel

Adrian Łańcucki

University of Wroclaw, Poland

Scientific Organizers

Nihat Ay

Max Planck Institute for Mathematics in the Sciences (Leipzig), Germany

Marc Toussaint

Stuttgart University, Germany

Administrative Contact

Marion Lange

Stuttgart University / TU Berlin, Germany Contact via Mail