journal article·2018-05-21Open access

Self-optimized construction of transition rate matrices from accelerated atomistic simulations with Bayesian uncertainty quantification

Thomas D. Swinburne, Danny Perez

Physical Review Materials· Volume 2· Issue 5

Abstract

A massively parallel method to build large transition rate matrices from temperature-accelerated molecular dynamics trajectories is presented. Bayesian Markov model analysis is used to estimate the expected residence time in the known state space, providing crucial uncertainty quantification for higher-scale simulation schemes such as kinetic Monte Carlo or cluster dynamics. The estimators are additionally used to optimize where exploration is performed and the degree of temperature acceleration on the fly, giving an autonomous, optimal procedure to explore the state space of complex systems. The method is tested against exactly solvable models and used to explore the dynamics of C15 interstitial defects in iron. Our uncertainty quantification scheme allows for accurate modeling of the evolution of these defects over timescales of several seconds.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Machine Learning in Materials Science · Nuclear Materials and Properties · Fusion materials and technologies

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

References (38)

Show source-provided references
1. 10.1103/PhysRevLett.78.390810.1103/PhysRevLett.78.3908 ↗
2. 10.1103/PhysRevB.57.R1398510.1103/PhysRevB.57.R13985 ↗
3. 10.1063/1.48157610.1063/1.481576 ↗
4. 10.1016/S1574-1400(09)00504-010.1016/S1574-1400(09)00504-0 ↗
5. 10.1146/annurev-matsci-071312-12161610.1146/annurev-matsci-071312-121616 ↗
6. 10.1103/PhysRevE.84.04670410.1103/PhysRevE.84.046704 ↗
7. 10.1080/0026897021016269110.1080/00268970210162691 ↗
8. 10.1016/j.ymeth.2010.06.00210.1016/j.ymeth.2010.06.002 ↗
9. 10.1063/1.481232310.1063/1.4812323 ↗
10. 10.1021/ct500871810.1021/ct5008718 ↗
11. 10.1103/PhysRevLett.114.10550310.1103/PhysRevLett.114.105503 ↗
12. 10.1146/annurev-chembioeng-080615-03360810.1146/annurev-chembioeng-080615-033608 ↗
13. 10.1103/PhysRevLett.74.110.1103/PhysRevLett.74.1 ↗
14. 10.1021/ct049900m10.1021/ct049900m ↗
15. 10.1063/1.275315310.1063/1.2753153 ↗
16. 10.1063/1.488072110.1063/1.4880721 ↗
17. 10.1063/1.498493210.1063/1.4984932 ↗
18. 10.1063/1.493097610.1063/1.4930976 ↗
19. S. N. Ethier · 2009
20. 10.1103/RevModPhys.62.25110.1103/RevModPhys.62.251 ↗
21. 10.1103/PhysRevLett.120.13550310.1103/PhysRevLett.120.135503 ↗
22. 10.1063/1.132967210.1063/1.1329672 ↗
23. 10.1103/PhysRevB.87.21410610.1103/PhysRevB.87.214106 ↗
24. 10.1016/S0022-2496(02)00028-710.1016/S0022-2496(02)00028-7 ↗
25. 10.1098/rspa.1946.005610.1098/rspa.1946.0056 ↗
26. 10.1063/1.354182310.1063/1.3541823 ↗
27. 10.1021/acs.jctc.5b0091610.1021/acs.jctc.5b00916 ↗
28. 10.1006/jcph.1995.103910.1006/jcph.1995.1039 ↗
29. 10.1103/PhysRevLett.97.17020110.1103/PhysRevLett.97.170201 ↗
30. 10.1016/0021-9991(75)90060-110.1016/0021-9991(75)90060-1 ↗
31. 10.1038/nmat128610.1038/nmat1286 ↗
32. 10.1103/PhysRevLett.108.02550110.1103/PhysRevLett.108.025501 ↗
33. 10.1016/j.jnucmat.2010.05.01710.1016/j.jnucmat.2010.05.017 ↗
34. 10.1088/0953-8984/25/12/12540210.1088/0953-8984/25/12/125402 ↗
35. 10.1016/j.jnucmat.2014.02.01210.1016/j.jnucmat.2014.02.012 ↗
36. 10.1016/j.scriptamat.2014.10.03310.1016/j.scriptamat.2014.10.033 ↗
37. D. J. Wales · 2003
38. 10.1016/j.jsc.2013.09.00310.1016/j.jsc.2013.09.003 ↗