Speaker:   Dr. Mike Preuss, WWU Muenster

Abstract:   To plan the syntheses of small organic molecules, chemists use retrosynthesis, a technique in which target molecules are recursively transformed into increasingly simpler precursors. Computer-aided retrosynthesis would be a valuable tool but at present it is slow and provides results of unsatisfactory quality. Here we use Monte Carlo tree search to discover retrosynthetic routes. We combined Monte Carlo tree search with an expansion policy network that guides the search, and a filter network to pre-select the most promising retrosynthetic steps. These deep neural networks were trained on essentially all reactions ever published in organic chemistry. Our system solves for almost twice as many molecules, thirty times faster than the traditional search method based on extracted rules and hand-designed heuristics.

Short Bio:   Mike Preuss is Research Associate at ERCIS, University of Muenster, Germany. Previously, he was with the Chair of Algorithm Engineering at TU Dortmund, Germany, where he received his PhD in 2013. His research interests focus on the field of evolutionary algorithms for real-valued problems, namely on multimodal and multiobjective optimization, and on computational intelligence methods for computer games, especially in procedural content generation (PGC) and realtime strategy games (RTS).

He is also interested in applying the game AI techniques to engineering problems, e.g., chemical retrosynthesis. Since 2016, he is involved in the PropStop project that deals with the detection of Propaganda on social media.

He is associate editor of the IEEE ToG (Transactions on Games) and advisory board member of Springer's Natural Computing book series and has been member of the organizational team of several conferences in the last years, in various functions, as general co-chair, proceedings chair, competition chair, workshops chair, notably also as PC co-chair for CIG 2018.