Fixed-sample optimization using a probability density function
- Lawrence Berkeley Lab., CA (United States)
We consider the problem of optimizing parameters in a trial function that is to be used in fixed-node diffusion Monte Carlo calculations. We employ a trial function with a Boys-Handy correlation function and a one-particle basis set of high quality. By employing sample points picked from a positive definite distribution, parameters that determine the nodes of the trial function can be varied without introducing singularities into the optimization. For CH as a test system, we find that a trial function of high quality is obtained and that this trial function yields an improved fixed-node energy. This result sheds light on the important question of how to improve the nodal structure and, thereby, the accuracy of diffusion Monte Carlo.
- DOE Contract Number:
- AC03-76SF00098
- OSTI ID:
- 560357
- Report Number(s):
- CONF-970443-; TRN: 97:005895-0274
- Resource Relation:
- Conference: 213. national meeting of the American Chemical Society, San Francisco, CA (United States), 13-17 Apr 1997; Other Information: PBD: 1997; Related Information: Is Part Of 213th ACS national meeting; PB: 2904 p.
- Country of Publication:
- United States
- Language:
- English
Similar Records
Communication: Toward an improved control of the fixed-node error in quantum Monte Carlo: The case of the water molecule
Quantum Monte Carlo with variable spins: Fixed-phase and fixed-node approximations