Another title: Computational Materials Design: Is this some mega-hype, or are our jobs really in danger?
From the MIT news release:
The same computer methods used by online sales sites to suggest books to customers can help predict the crystal structures of materials, MIT researchers have found.
These structures are key to designing new materials and improving existing ones, which means that everything from batteries to airplane wings could be influenced by the new method. [...]
Using a technique called data mining, the MIT team preloaded the entire body of historical knowledge of crystal structures into a computer algorithm, or program, which they had designed to make correlations among the data based on the underlying rules of physics.
Harnessing this knowledge, the program then delivers a list of possible crystal structures for any mixture of elements whose structure is unknown. The team can then run that list of possibilities through a second algorithm that uses quantum mechanics to calculate precisely which structure is the most stable energetically -- a standard technique in the computer modeling of materials.
The Nature Materials paper from Gerbrand Ceder's group is here [subscription required].
2 Comments:
The prediction capability of any mining/learning algorithm is decided by the quality of training data.
I have not read the paper and so I cannot comment how good the predictions would be. But in general I can say that so far no machine learning algorithms have come to 100% accuracy on prediction.
Personally, I would say if correlations are obvious and the identification process can be broken down into a sequence of steps (algorithms), then it should be done.
Ah well I am biased I guess
no subscription, so couldnt read it. but, sounds rather similar to the Phd work of one of Gerd Cedar's students, Stefano Curtarolo (presently a faculty at Duke University's dept of mechanical and Materials Engg). His thesis can be downloaded from Cedar's webpage.
Again, not having read the paper, i dont really know whats been done there, but the last i heard was they were dealing with binaries, and were trying to extend coarse graining and data-mining of quantum calculations to ternary alloys.
Post a Comment