Microsoft lab to ‘supercharge molecular science’ with machine learning

21 Jul 2021

Image: © fusebulb/Stock.adobe.com

By using the latest advances in AI, the new Microsoft lab will look to better understand the nitty-gritty details of molecular properties and behaviours.

Microsoft Research is opening a lab in Amsterdam focused on advancing molecular simulation by using machine learning, quantum chemistry and quantum computing.

The lab will be led by Dr Max Welling, who specialises in computer science and machine learning. Welling is currently based at the University of Amsterdam and will be joining Microsoft Research in September.

He was hosted on the Microsoft Research podcast this week by Chris Bishop, a lab director of Microsoft Research in Europe, where the announcement was made.

“Molecules are basically everything around us, except for light and a few other forces that we can’t really see. Everything else is made of molecules and yet we don’t really understand them, we can’t really predict their properties,” said Welling.

“So, if we start to understand molecules better, then a number of applications become within reach. We can start to design better catalysts, for instance to help the hydrogen economy.”

This new lab in Amsterdam will use cloud-scale deep-learning technology to better understand how molecules behave, and predict and simulate properties that serve societal needs.

Through these simulations, the goal is to address issues related to the climate crisis, healthcare, drug discovery and sustainable materials.

Bishop leads the Cambridge Microsoft Research lab in the UK. The team there focuses on deep research in areas such as cloud infrastructure, confidential computing and machine intelligence.

“Building on foundational molecular modelling research, we aim to supercharge molecular science by using machine learning,” said Bishop about the new lab in Amsterdam.

“Just as it has already revolutionised fields like computer vision, speech recognition and natural language understanding, we believe machine learning is set to make a similar transformation in the field of molecular simulation.

“This has tremendous potential for real-world impact on some of the most important challenges facing society including climate change, drug discovery and understanding biology to help us treat disease.”

Sam Cox was a journalist at Silicon Republic covering sci-tech news

editorial@siliconrepublic.com