nach langer Pause haben wir wieder zwei spannende Vorträge, beide diesmal auf englisch:
Moritz Wolter von der Uni Bonn wird sein Paper zu “Complex gated recurrent neural networks” vorstellen:
Fourier methods have a long and proven track record as an excellent tool in data processing. During the talk we will discuss integration of Fourier methods into complex recurrent neural network architectures and look at their application to time series prediction tasks like power load forecasting (see https://arxiv.org/abs/1812.05645),
Mihail Luchian von der tarent solutions GmbH wird etwas über “Generative Art” erzählen:
Generating uniqueness – How generative art pushes the boundaries of intellectual property
Recent advancements in Machine Learning and Blockchain technology, have led to a renewed interest in Generative Art. The way in which these fields work and rely on randomness has inspired artists and coders to explore what it means for a piece of content to be unique and be posessed by someone. We believe the best way to illustrate such problems is by actually showing how one goes from an initial random number to an actual image and this is exactly what we will be doing in the talk!
The talk will start with an overall presentation of Generative Art and describe the recent influence that Machine Learning and Blockchain have had on this art form. Ample examples and references will, of course, be present. What emerged from these recent influences is a new and different way of thinking about digital content, where ownership is expressed through smart contracts on the blockchain or by secret knowledge of the initial seed used to generate the image. Throw in Open Source in the mix and it might just be impossible to make any more sense of intelectual property. As the presented material can be hard to grasp without an actual example, in the second part we will illustrate a typical workflow with which generative art is created. In the process, the audience will have the opportunity to explore different generation and structuring techniques, fundamental building blocks of the art form, like pseudorandom number generator, seeds, simple probability distributions, and the issues one often has to deal with.