Tübingen: MPI for Intelligent Systems Scientists among the winning team at world’s leading AI conference

Photo: unsplash.com/ Franck V.

Scientists from the Max Planck Institute for Intelligent Systems Tübingen won the Best Paper Award 2019 in collaboration with ETH Zurich, and Google Research Zurich for their research project “Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations”. A total of 750 contributions were accepted at this year's International Conference on Machine Learning (ICML) – one of the world's leading conferences in the field of machine learning.

"This award shows that European scientists are at the forefront of modern AI research," says Bernhard Schölkopf, Director at the Max Planck Institute for Intelligent Systems in Tübingen and one of the world's leading researchers in the field of machine learning.

In their research, the scientists feed a computer thousands of images showing colored objects of different shapes. The images consist of only a few pixels and are two-dimensional. From this large data set, the computer has to identify patterns.

Imagine a setup where a robot arm is being. In one task, it should pick up red squares from a table. With enough trials and data, a deep neural network can be trained for the task. However, the machine fails if the researcher suddenly puts a yellow triangle on the table and asks it to lift that out. The machine cannot interpret a new pattern as such. It cannot transfer what it has learned before. "That is one of the key problems in AI", says Stefan Bauer, Research Group Leader at the Max Planck Institute for Intelligent Systems. "We don't want to have to train a new Deep Net for every new color and every new shape each and every time."Ideally, the computer would come to the conclusion on its own, without a scientist having to train it again.  In conclusion, there is still enough research to do in the field of machine learning.

For more information please visit www.is.mpg.de

Baden-Württemberg International

Haus der Wirtschaft
Willi-Bleicher-Strasse 19
D-70174 Stuttgart
phone: (++49) 711 / 2 27 87 - 30
fax: (++49) 711 / 2 27 87 - 72