Showing 1–2 of 2 results for author: Marks, L D

Reply to "Local Filtering Fundamentally Against Wide Spectrum"

Authors: Jianwei Miao, M. C. Scott, Chien-Chun Chen, Chun Zhu, Edward R. White, Chin-Yi Chiu, B. C. Regan, Yu Huang, Laurence D. Marks

Abstract: After carefully studying the comment by Wang et al. (arXiv:1408.6420), we found it includes several mistakes and unjustified statements and Wang et al. lack very basic knowledge of dislocations. Moreover, there is clear evidence indicating that Wang et al. significantly misrepresented our method and claimed something that they actually did not implement. After carefully studying the comment by Wang et al. (arXiv:1408.6420), we found it includes several mistakes and unjustified statements and Wang et al. lack very basic knowledge of dislocations. Moreover, there is clear evidence indicating that Wang et al. significantly misrepresented our method and claimed something that they actually did not implement. △ Less

Submitted 2 September, 2014; originally announced September 2014.

Comments: 2 pages, 9 references

Enhancing structure relaxations for first-principles codes: an approximate Hessian approach

Authors: J. M. Rondinelli, B. Deng, L. D. Marks

Abstract: We present a method for improving the speed of geometry relaxation by using a harmonic approximation for the interaction potential between nearest neighbor atoms to construct an initial Hessian estimate. The model is quite robust, and yields approximately a 30% or better reduction in the number of calculations compared to an optimized diagonal initialization. Convergence with this initializer ap… ▽ More We present a method for improving the speed of geometry relaxation by using a harmonic approximation for the interaction potential between nearest neighbor atoms to construct an initial Hessian estimate. The model is quite robust, and yields approximately a 30% or better reduction in the number of calculations compared to an optimized diagonal initialization. Convergence with this initializer approaches the speed of a converged BFGS Hessian, therefore it is close to the best that can be achieved. Hessian preconditioning is discussed, and it is found that a compromise between an average condition number and a narrow distribution in eigenvalues produces the best optimization. △ Less

Submitted 13 November, 2006; v1 submitted 15 August, 2006; originally announced August 2006.

Comments: 9 pages, 3 figures, added references, expanded optimization section

Journal ref: Comput. Mater. Sci. 40 (2007) 345-353