Tuesday, September 25, 2018 at 4:00pm to 5:15pm
Physical Sciences Building, 401
245 East Avenue
Professor Shoucheng Zhang, Department of Physics, Stanford University, Stanford, CA, will present seminar. Professor Eun-Ah Kim, host.
Seminar Title: Al for Material Discovery
Abstract: Exciting advances have been made in artificial intelligence (AI) during the past decades. Among them, applications of machine learning (ML) and deep learning techniques brought human-competitive performances in various tasks of fields, including image recognition, speech recognition and natural language understanding. Even in Go, the ancient game of profound complexity, the AI player already beat human world champions convincingly with and without learning from human. Atomism - the idea that the world is made out of atoms, and their organizational chart in terms of the periodic table - is certainly one of the greatest scientific achievements of human intelligence. In this work, we show that our unsupervised machines (Atom2Vec) can discover the basic properties of atoms by themselves from the extensive database of known compounds and materials. These learned properties are represented in terms of high dimensional vectors, and clustering of atoms in vector space classifies them into meaningful groups consistent with human knowledge. We use the atom vectors as basic input units for neural networks and other ML models designed and trained to predict materials properties, with significant accuracy.