Air pollution is still a major problem in cities all over the world, therefore a short term forecast of pollutant loads across critical regions is of interest for decision makers to take appropriate measures. In this thesis, a sequence-to-sequence recurrent neural network has been implementented and trained to perform a 48h pollutant load forecast for measuring stations across North-Rine-Westfalia on the basis of measured pollutant load and weather data. The network is based on Gated Recurrent Units (GRU) and extended with an attention mechanism. The network outperforms a baseline neural network and the attention mechanism provides interpretability of the results. The thesis is developed in cooperation with the Fraunhofer Heinrich-Hertz-Institute as part of project SAUBER.
The collective behaviour of groups of animals emerges from interaction between individuals. Understanding these interindividual rules has always been a challenge, because the cognition of animals is not fully understood. Artificial neural networks in conjunction with attribution methods and others can help decipher these interindividual rules. In this thesis, an artificial neural network was trained with a recently proposed learning algorithm called Soft Q Imitation Learning (SQIL) on a dataset of two female guppies. The network is able to outperform a simple agent that uses the action of the most similar state in defined metric and also is able to show most characteristics of fish, at least partially, when simulated.