Source author record

Brendan Gavin

Brendan Gavin appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Numerical Analysis cond-mat.mes-hall cond-mat.mtrl-sci Machine Learning math.NA physics.chem-ph physics.comp-ph Social and Information Networks

Catalog footprint

What is connected

3works

8topics

3close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2016arXiv

Enhancing the Performance and Robustness of the FEAST Eigensolver

The FEAST algorithm is a subspace iteration method that uses a spectral projector as a rational filter in order to efficiently solve interior eigenvalue problems in parallel. Although the solutions from the FEAST algorithm converge rapidly in many cases, convergence can be slow in situations where the eigenvalues of a matrix are densely populated near the edges of the search interval of interest, which can be detrimental to parallel load balancing. This work introduces two methods that allow one to improve the convergence robustness of the FEAST algorithm in these situations without having to increase the amount of computation. Selected numerical examples are presented and discussed

preprint2015arXiv

Large Enforced Sparse Non-Negative Matrix Factorization

Non-negative matrix factorization (NMF) is a common method for generating topic models from text data. NMF is widely accepted for producing good results despite its relative simplicity of implementation and ease of computation. One challenge with applying NMF to large datasets is that intermediate matrix products often become dense, stressing the memory and compute elements of a system. In this article, we investigate a simple but powerful modification of a common NMF algorithm that enforces the generation of sparse intermediate and output matrices. This method enables the application of NMF to large datasets through improved memory and compute performance. Further, we demonstrate empirically that this method of enforcing sparsity in the NMF either preserves or improves both the accuracy of the resulting topic model and the convergence rate of the underlying algorithm.

preprint2013arXiv

Non-linear eigensolver-based alternative to traditional SCF methods

The self-consistent procedure in electronic structure calculations is revisited using a highly efficient and robust algorithm for solving the non-linear eigenvector problem i.e. H({ψ})ψ = Eψ. This new scheme is derived from a generalization of the FEAST eigenvalue algorithm to account for the non-linearity of the Hamiltonian with the occupied eigenvectors. Using a series of numerical examples and the DFT-Kohn/Sham model, it will be shown that our approach can outperform the traditional SCF mixing-scheme techniques by providing a higher converge rate, convergence to the correct solution regardless of the choice of the initial guess, and a significant reduction of the eigenvalue solve time in simulations.