Artificial intelligence and the climate debate are the topics at this year's "Day of Mathematics"

As Global Head of Data Science at Merck KGaA, Dr. Helmut Linde is responsible for harnessing innovations from machine learning and artificial intelligence in the business. With his background as a physicist and mathematician, Linde initially spent ten years in the software industry, where he built up a consulting business around data science projects for a leading corporation, until he took on his current position in 2017. In his presentation "Making Molecules Smarter - Artificial Intelligence in Use at Merck" he will address the increasingly important role of machine learning and artificial intelligence in the chemical and pharmaceutical industries. From researching new materials and active ingredients, to optimizing production and the supply chain, to targeted sales management, algorithms are helping to accelerate growth and reduce costs. Combining computational chemistry with machine learning in particular is proving extremely fruitful. Deep Learning also promises to open up completely new fields of application, for example in the diagnosis of diseases. At the same time, we are seeing more and more clearly the limitations of the artificial neural networks currently in use. Will a new generation of AI algorithms overcome these limitations and enable entirely new applications?
Our second speaker Prof. Dr.-Ing. Rupert Klein is a professor of computational fluid dynamics at Freie Universität Berlin and was head of the Data and Computation department at the Potsdam Institute for Climate Research (PIK) from 1997 to 2007. In 2003, he received the Gottfried Wilhelm Leibniz Prize, the highest research funding award in Germany. His work deals with mathematical models and numerical methods relevant for climate prediction. He is an associate member of the Berlin-Brandenburg Academy of Sciences. Using three examples, Prof. Dr.-Ing. Klein will show in his talk "How Mathematics Helps Structure the Climate Debate" that mathematics is more than just a provider of methods for solving climate model equations. The first example demonstrates that mathematics can provide systematic guidelines for applicability and validity of simplified models. The second example describes recent developments in methods of mathematical time series analysis that are needed to describe climate change correctly. The third example shows how the ability of mathematicians to define terms can provide clarity in interdisciplinary discussions with social sciences and economics.