The first one where I started to prepare some notes was ICLR 2017. You can find pictures of posters and some notes here.

]]>

Mo 12c.t.-14 Lecture in A301, Th. 12c.t.-14 Recitation in A301. See also the see also the entry in “campus verwaltung”.

The course will provide you with the theoretical and practical knowledge of a subset of Machine Learning mostly suitable for robotic applications. We start with the basics of supervised learning, model selection etc, then turn to unsupervised learning and spend the most time on reinforcement learning. The exercised will help you to get Hands On with the methods and deepen your understanding.

- Lecture: Slides-1a, Slides-1b, Background reading: C.M. Bishop
*Pattern Recognition and Machine Learning*, Chap. 3 - Lecture: Slides-2, Slides-2 (4 on 1), Background reading: C.M. Bishop
*Pattern Recognition and Machine Learning*, Chap. 1, 2, 4 - Lecture: Slides-3, Slides-3 (4 on 1), Background reading: C.M. Bishop. Chap. 4
- Lecture: Slides-4, Slides-4 (4 on 1), Linkage Clustering IPython Background reading: S. Shalev-Shwartz and S. Ben-David Understanding Machine Learning
- Lecture: Slides-5, Slides-5 (4 on 1), Background reading: A. Ghodsi Dimensionality Reduction Tutorial, P. Vincent et al Stacked Denoising Autoencoder
- Lecture: Slides-6, Slides-6 (4 on 1), Background reading: Sutton and Barto Reinforcement learning for the next few lectures
- Lecture: Slides-7, Slides-7 (4 on 1) and Slides-7b (David Silvers slides 3)
- Lecture: Slides-8a and Slides-8b (David Silvers slides 4 and 5)
- Lecture: Slides-9, (David Silvers slides 6)
- Lecture: Slides-10, (David Silvers slides 7)
- Lecture: Slides-11, (David Silvers slides 8)
- Lecture: Slides-12, collection of resources, mostly on blackboard see PGPE-paper

Sheet | Due Date | Additional Information | Solution (code) |

exercise sheet 1 | 27.04.2017 | Data: diabetes.txt, Python cheat sheet, Some matplotlib tutorial, Otherwise google helps |
solution 1 code |

exercise sheet 2 | 04.05.2017 | Data: diabetes.txt | solution 2 code |

exercise sheet 3 | 11.05.2017 | Updated on 8th.May! 3 exercises. | |

exercise sheet 4 | 18.05.2017 | Notebook to start with. | solution 4 code |

exercise sheet 5 | 01.06.2017 | Data files: 4Wheeled, Sliderwheelie-pimax-env.x, and v2 | solution 5 code |

exercise sheet 6 | 01.06.2017 | Data files as for exercise 5 | solution 5 + 6 code |

exercise sheet 7 | 22.06.2017 | Code basis gridworld.zip, or gridworld3.zip (for Python 3) You need tkinter and python-tk installed. |
solution 7 as zip |

exercise sheet 8 | 22.06.2017 | Code basis qlearning.zip, or qlearning3.zip (for Python 3) | solution |

exercise sheet 9 | 06.07.2017 | Notebook: exercise9.ipynb | |

Final project | 03.08.2017 | possible solution |

In particular for mechanical systems which are governed by the laws of physics, if we can learn the actual equations then accurate predictions outside of the domain of the available data can be made, i.e. we are able to extrapolate. Our current approach uses a modified feed-forward network with sine, cosine, identity and multiplicative units (with two terms) and a particular regularization and model selection scheme.

Preprint paper: G. Martius and C. H. Lampert. Extrapolation and learning equations. *arXiv* 1610.02995. [ arXiv ]

]]> Check out the videos and the papers below.

Papers: G. Martius, R. Hostettler, A. Knoll, and R. Der. Compliant control for soft robots: emergent behavior of a tendon

G. Martius, R. Hostettler, A. Knoll, and R. Der. Compliant control for soft robots: emergent behavior of a tendon driven anthropomorphic arm. In Proc. IEEE Int. Conf. on Intelligent Robots and Systems (IROS), 2016. in press. [ bib | Supplementary page (videos) | pdf ]

]]>

Since 10 years we organize in Leipzig and Halle (Germany) an annual workshop called HAL on theory and application of Haskell. Big thanks to Johannes Waldmann for his major efforts in Organization.

See webpage for more details.

4. und 5. Dezember 2015, HTWK Leipzig

]]>

Paper: R. Der and G. Martius. Novel plasticity rule can explain the development of sensorimotor intelligence. *Proceedings of the National Academy of Sciences*, 2015. . [ bib | DOI | http | arXiv ]

See also the Supplementary page.

]]>

Georg Martius

Am Campus 1,

3400 Klosterneuburg,

Austria

Seite: http://georg.playfulmachines.com

Email:

Haftungsausschluss

Monday 20th July 2015, 10:00 – 13:00

In the workshop “Information Theory in Artificial Life” we will discuss how Information Theory can be used to generate, motivate, understand and quantify the behaviour and other processes in artificial agents and life-like systems. Information Theory provides a language to express the required concepts and quantities in a general way, allowing to transfer them between different domains. For further details and information on how to submit and participate see our website. If you have any questions please send us an email at itialife@gmail.com.

Organised by: Georg Martius, Christoph Salge, Keyan Ghazi-Zahedi, and Daniel Polani

]]>New York, 30th of July 2014

]]>Paper: G. Martius and E. Olbrich. Quantifying emergent behavior of autonomous robots. Entropy, 17(10):7266, 2015. [ bib | DOI | http ]

See also the Supplementary page.

]]>