3 ms·
Nice trick, but it’s almost useless. We have been using for decades integer programming to explore all the possible permutations with hard constraints that inc
by whatever1 3y ago
Nice trick, but it’s almost useless.
We have been using for decades integer programming to explore all the possible permutations with hard constraints that include manufacturability.
Their references list is lacking, to say the least.
- rsfern 3y agoit would be great if you could provide a bit more detail to your criticism, especially pointers to relevant literature you think they missed Frankly I don’t really see how some discrete optimization thing solves the problem this paper is addressing, because evaluation of the constraints (i.e., thermodynamic stability of a crystal) is one of the most computationally challenging aspects of the problem
- rossjudson 3y agoThe paper (and graphics inside) refer to 48,000 materials that have been discovered by previous computational methods. Is this what you mean? Looks like the contribution here is an order of magnitude increase in high probability stable materials.