- Title: AlphaGo in Chemistry – Solving Real-World Problems by Means of Game AI Methods
- Speaker: Mike Preuss, Universiteit Leiden
- Time and date: 4pm to 5pm, May 22, 2019
- Room: 3.01, Bancroft Road Teaching Rooms, Mile End campus
On Tuesday 22nd May 2019 the Game AI Group will host a seminar by Mike Preuss from Universiteit Leiden. All welcome (especially students), no pre-booking required. The seminar will be followed by drinks in The Hub.
Abstract
Monte Carlo Tree Search (MCTS) and Deep Neural Networks (DNN) in combination have pushed the limits of what artificial intelligence (AI) can do in areas where humans have been perceived as dominant over machines, as for the game Go. However, so far this has been limited to domains with simple and known rules, such as board games, where perfect world knowledge and cheap environment simulators are available. Unfortunately, this is usually not the case for real-world problems, usually we cannot obtain the rules themselves but only acquire knowledge from interactions with the world or simulation.
Chemical retrosynthesis (you know the product, but not how to get there) is one of such real-life domains with highly non-trivial, partially unknown rules. We show that this problem can very effectively be tackled with MCTS constrained by DNNs that learn from essentially the complete history of organic chemistry, as described in a recent Nature paper [1]. Our algorithm produces plans which human chemists could not tell apart from established plans taken from the literature. But it does not have to end here. We attempt to generalize the approach in order to make it applicable to other research areas as well.
Bio
Mike Preuss is Assistant Professor at LIACS, the computer science institute of Universiteit Leiden in the Netherlands. Previously, he was with ERCIS (the information systems institute of WWU Muenster, Germany), and before with the Chair of Algorithm Engineering at TU Dortmund, Germany, where he received his PhD in 2013. His main research interests rest on the field of evolutionary algorithms for real-valued problems, namely on multimodal and multiobjective optimization, and on computational intelligence and machine learning methods for computer games, especially in procedural content generation (PGC) and realtime strategy games (RTS).
References
[1] Segler, M. H., Preuss, M., & Waller, M. P. (2018). Planning chemical syntheses with deep neural networks and symbolic AI. Nature, 555(7698), 604. https://www.nature.com/articles/nature25978