Machine-learning algorithm aims to accelerate materials discovery - Argonne National Laboratory

[Bill USA]

... Our government foots the bill for the research and produces the basic knowledge underpinning new technologies/products. Then private sector firms (operated by those 'by-their-Bootstraps' entrepreneurs) step in, once all the research is done, and commercialize it, making billions of $s. They then form Political Action Committees funding GOP imbeciles and publishing propaganda saying Government doesn't create jobs.


http://www.anl.gov/articles/machine-learning-algorithm-aims-accelerate-materials-discovery

July 16, 2013

A research team led by Argonne Leadership Computing Facility computational chemist Anatole von Lilienfeld is developing an algorithm that combines quantum chemistry with machine learning (artificial intelligence) to enable atomistic simulations that predict the properties of new materials with unprecedented speed. From innovations in medicine to novel materials for next-generation batteries, this approach could greatly accelerate the pace of materials discovery, with high-performance computing tools offering important assistance, or even full-blown alternatives, to time-consuming laboratory experiments.

The team applies statistical learning techniques to the problem of inferring solutions of the electronic Schrödinger equation, one of the most important equations in atomistic simulation using quantum chemistry. Conventionally, this differential equation is solved with an algorithmic method that iteratively approaches the electronic wavefunction in which the potential energy of the system is minimal. The computational effort for this task is significant, and has historically resulted in a strong presence of quantum chemists at high-performance computing centers around the world. While this deductive approach is perfectly valid, one has to restart the iterations again every time the chemical composition changes. Von Lilienfeld’s approach attempts to infer the solution for a newly composed material instead, provided that a sufficiently large number of examples have been used for training.

Machine-learning techniques require very large amounts of data in order to infer solutions for interpolating scenarios with satisfying accuracy. Many thousands, if not more, data points are required. At such scale, it rapidly becomes unfeasible to obtain the data from experiment, and instead the team relies on supercomputers to simulate such large numbers of materials. Von Lilienfeld and his colleagues have made significant advances by running simulations on Intrepid supercomputer through an in-house discretionary program that awards start-up time to researchers with a demonstrated need for leadership-class computing resources.


von Lilienfeld, O. A. (2013), First principles view on chemical compound space: Gaining rigorous atomistic control of molecular properties. Int. J. Quantum Chem., 113: 1676–1689. doi: 10.1002/qua.24375