The full configuration interaction quantum Monte Carlo (FCIQMC) method, developed by Alavi and coworkers, allows one to compute FCI energies for considerably larger systems than is possible by conventional FCI methods.  However, it is very computationally expensive.  The semistochastic quantum Monte Carlo (SQMC) [1] achieves three orders of magnitude gain in efficiency by performing a deterministic projection in a small but important part of Hilert space and a stochastic projection in the rest of the space and by employing an optimized trial wave function.  In addition, the already small 'initiator bias' is further reduced.  Further improvements to the SQMC method and its application to the G2 get of molecules is discussed.[1] F. R. Petruzielo,1, A. A. Holmes, Hitesh J. Changlani, M. P. Nightingale, and C. J. Umrigar, Phys. Rev. Lett., 109, 230201 (2012)

Semistochastic quantum Monte Carlo method and applications

246th National Meeting (2013)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.