Machine Learning, Neural Networks Probe the Interactions of Water Molecules

An article by Pacific Northwest National Laboratories (PNNL) on shares how machine learning algorithms, the basis of neural networks, are bringing new scientific discoveries closer to reality.

Highlighted in the piece is the work of the ExaLearn Co-Design Center in the US Department of Energy’s Exascale Computing Project. ExaLearn focuses on machine learning technologies. The center has as one of its primary aims the development of artificial intelligence (AI) technologies that can design new chemicals by learning from massive sets of data. Machine learning is a subset of AI.

The article expresses how researchers in the PNNL-led study trained the neural network to probe the structural patterns of clusters of water molecules using just one-tenth of a database of over 5 million water cluster minima. The researchers then used what’s called graph theory to analyze the structural patterns of the water networks.

Read more.