Seminar về chủ đề “AI for Matters”

2/22/2019 3:53 AM

Trình bày: GS. Trần Thế Truyền, Deakin University.

Abstract: We are living in a data-driven world. The ground for scientific discovery is shifting towards the fourth paradigm, which is founded on the premise that much hidden knowledge can be uncovered by analyzing massive data sets generated by physical processes. A parallel has also been seen in modern AI. Fueled by recent advances in deep learning, AI thrives on human and machine-generated data, breaking records in numerous data-intensive fields it touches. As a general purpose technology, AI could also thrive on scientific data too. This has resulted in a considerable effort to capitalize AI for accelerating the progress of science in recent years, especially in the analysis and discovery of new molecules and materials, thanks to the public availability of datasets curated by the scientific community. This talk covers the recent developments in this young promising playground, starting with a brief introduction to modern AI, then diving into exploration of the molecular space in chemistry, pharmaceutical drugs and genomics. The talk then discusses representation of materials that facilitates functional mapping between structure, property and performance.

Speaker info: Dr Truyen Tran is currently leading a research team on deep learning technologies at Applied AI Institute, Deakin University, Australia. He started his scientific inquiry in physics, then ventured into computer science, and subsequently settled in AI. His work has been well-received with multiple best-paper awards. One of his long-term interests is uncovering the hidden patterns of nature using AI facilities such as machine learning and automated reasoning. His current work includes characterizing the molecular and material space and deriving data-driven methods for automated discovery. Tran holds a Bachelor in Science from the University of Melbourne in 2001 and a PhD in Computer Science from Curtin University in 2008. After graduation, he spent a year in the industry, then 3 years as a postdoc at Curtin University before moving to Deakin University, where he is currently an associate professor in AI.

Link sự kiện: