Events

In this talk I’ll describe the use of artificial neural networks (ANN) for quantifying X-ray fluorescence (XRF) measurements. The main idea of this talk is to give an overview of the process needed to generate a model that can then be applied to a specific problem. 

In XRF, a sample is excited with X-rays and the resulting characteristic radiation is detected to determine elements quantitatively and qualitatively. This is traditionally done in several time-consuming steps. I’ll show the possibilities and problems of using a neural network to realise a “one-click” quantification. This includes generating training data using Monte Carlo simulation and augmenting the existing data set with an ANN to generate more data.  The search for the optimal hyperparameters, manually and automatically, is also described. For the case presented, we were able to train a network with a mean absolute error of 0.1% by weight for the synthetic data and 0.7% by weight for a set of experimental data obtained with certified reference materials. 

Bio:
Dr. Martin Radtke is the beamline scientist at BAMline. BAMline is the experimental station of the Federal Institute for Materials Research and Testing (BAM) at BESSY II in Berlin. His work focuses on the further development and application of X-ray fluorescence based techniques. The main applications are in material science, e.g. in the investigation of corrosion, and in archaeometry, with a focus on the investigation of archaeological gold objects. Martin graduated with a degree in physics from the University of Hamburg, where he also received his PhD in physics. He then spent three years in Brazil, where he worked as a scientist at the fluorescence beamline of the LNLS in Campinas and at the Pelletron accelerator of the University of Sao Paulo. His interest in machine learning is in anything that can make beam time more efficient and allow to get the most out of the data, e.g. automatic alignment and data analysis techniques.